I don't think I have a "burnout", but LLMs are really exhausting due to amount of pressure they generate. No one is really pushing me to increase my workload, but at every moment there is always something ready, done by my clankers or clankers of other people that I could be unblocking. In the past (before LLMs) it was already hard to keep up, but now it feels like there's 10x more things waiting at any given time, and there could be 10x more if everyone just "optimized" and streamlined processes fed the AI even more tasks in parallel faster. It just being a bottleneck of everything, all the time is tiring...
I am happy about all the little side-projects, and ideas it help my realize, and I enjoy exploring this new world, but I've noticed LLMs feed my unhealthy "don't want to take a break and waste time being idle" mindset, and I need to correct it.
W.r.t. article's main complain - I think the similar thing happened due to factory manufacturing automation. What used to be a varied skillful craft in a shop became standing in a single place of an assembly line doing the exact same thing whole day. LLM took away the more creative and variable part of the work, and left the repetitive QA rubber-stamping. Probably some of the mitigations used back then could be rediscovered today.
I'm getting so many requests to review LLM-generated documents - planning docs, docs intended for end-users, project docs, business plan docs. A team member sent me a zip file with about 30 LLM generated documents in it the other day and asked if I could review them right away. And a lot of it was just repetition and/or stuff that was just out of left field, made-up, hallucinated stuff. They're able to generate this stuff way faster than we can review. It used to be that it would take a significant part of a day for a project manager to come up with a planning doc - now they can generate one in a few minutes and send it out for review. It's just really tiring.
The only way to even start to counter that is to make it a firm company policy that if you use an LLM to hallucinate any documents you absolutely must thoroughly review them yourself before you send them to anybody else, and that you are still responsible for the quality of LLM-generated content.
Getting an LLM to vomit out a bunch of documents and sending them straight to another colleague is absolutely unacceptable behaviour.
This is going to just run up against the insanity that is tokenmaxxing every moment of the day. When people are incentivized (upon pain of firing) to get the LLM to vomit out as much as possible, they're hardly going to stop and ponder if schlepping the slop over the wall is acceptable if the alternative is a pink slip.
We employees need to remember that most software projects fail. So the work we produce should have lower value than we give it.
We also need to be motivated to stay in our jobs.
Most developers like their projects and value their work. But the chances are that it's for nothing.
Many developers know they work on bad products (gambling industry, military, surveillance, whatever) and so it's here that they focus on their technologies, tools and frameworks rather than the work they produce.
"Agentic engineering" for example.
Id be curious to see what and how Googlers are doing with their 20% time.
Don't know if you are serious, but why become part of the problem?
Why not just review a single document quickly, find an error which invalidates the document, and send it back saying "Policy paper 1 mentions X as being on the business plan for Y, it's not on the plan, please can you fix."
Unless you can write a good-sounding reason why it's on them to review a LLM output before sending it to you, they will outsources this reviewing to you, and it's a lot of reviewing.
> I'm getting so many requests to review LLM-generated documents
That's the other nightmare of AI slop. So easy to generate endless content. Who will review?
Just today the boss request I review slides for a presentation. But it's all AI slop, generated from querying tickets and docs and who knows what. It's mostly sort of correct but also plenty misleading and incorrect. So now I have to fact check all this slop which will take hours (even with my AI assistance) and rewrite most of it.
If AI didn't exist, he would've had to do the research to generate the content and it would be 99% correct and I could just give a few notes of feedback in 5 minutes. But with the asymmetric AI workload, he can generate it in 5 minutes and I get to spend 3 hours correcting.
Seems like for such requests it's necessary to get some proof of work: require a meeting where for every artifact they sent you to review, they briefly explain the gist and point out the motivation for creating the artifact.
I got this problem with my own employees, LLM are fine, but lazy slop is not permitted. Current idea is to have a clear "best practice" template for most of the research/specs/problem definition they submit and it reduced the slop to a manageable level.
But this might work in a smaller company where the management is reading and is strict about these things.
Wait, what? I thought everyone agrees that modern models post September 2025 (or whenever Opus or whatever 5.6789 was released) do not hallucinate, make things up, contradict themselves and can review their own output into perfection regardless of task, goal or context???? /s
In general I think from the coding side they're more robust now. However, people generating docs are maybe not as experienced with how to prompt in ways that avoid having the LLM tell you what you want to hear. I think this is still a pitfall that can easily be fallen into. Those of us who are doing LLM-assisted coding for the last couple of years are more aware of this now. Those who are planning/management folks are still kind of susceptible depending on how much experience they've had dealing with LLMs.
They do, just less. To the degree of being usable, as long as there are guardrails and they're used responsibly. For example, if there's code being output, there should be type checking and compilation, as well as code tests that prove that it works or that it doesn't - seeing how abysmal code coverage is in most of the projects I've seem, for whatever reason people thought that they didn't really need it much. They were wrong.
This also implies you need SOTA models on max reasoning.
> make things up
Same as above. Ideally you'd give them some way to verify their claims, like web search or browsing and referencing docs, Jira tickets etc., basically improve the signal to noise ratio.
> contradict themselves
They do so way less than before, as long as the above is true.
> can review their own output into perfection
They are pretty good at reviewing things, especially if you make them do adversarial review! It will never be perfect, but can be close in quality to human output (e.g. the code they produce, when used properly and with intent, is better than the code I've seen many developers write and ship before LLMs were a thing).
This also more or less scales with how much compute you give them - three parallel review agents will turn one output artifact into something good with higher confidence than two, and definitely better than with no review. There's a cost vs quality balance and it seems that all those xhigh and max reasoning modes are still geared way too much towards cost, instead of quality. So you have to make up for that shortcoming yourself.
> regardless of task, goal or context????
Garbage in, garbage out. I won't be an asshole and say that you're holding it wrong, nor will I say that anyone should listen to the claims marketing AI (absolutely delusional takes, meant to attract investors), but we're slowly getting to a better position in regards to LLMs, year by year.
It's just a shame that the peak of inflated expectations hit while the technology still hasn't fully plateaued and reached whatever its ceiling is.
I probably also shouldn't ignore the fact that some people will not care about any of it and send AI generated slop verbatim and to an outside observer there's no way to easily tell apart the difference between the two, unless you make a technical report contain exact references to where the data is sourced from, for example (and then either verify the references yourself, or make another agent do it).
> I think the similar thing happened due to factory manufacturing automation. What used to be a varied skillful craft in a shop became standing in a single place of an assembly line doing the exact same thing whole day.
I had to think of the factory scenes in Charlie Chaplin's Modern Times. The author's feeling is basically the main idea of the sketches, i.e. humans having to follow the pace of the machines instead of the other way around.
Reverse centaurs are nothing new. Ask any worker movement from the last centuries.
> LLM took away the more creative and variable part of the work, and left the repetitive QA rubber-stamping
“I wanted a machine to do the dishes so I could concentrate on my creative work, and all I got was a machine to do my work so I’m left to wash the dishes.”
> I don't think I have a "burnout", but LLMs are really exhausting due to amount of pressure they generate. No one is really pushing me to increase my workload, but at every moment there is always something ready, done by my clankers or clankers of other people that I could be unblocking.
I see a different type of pressure: I'm at a company that still is requiring everyone use LLMs with token leaderboards, time-spent measurements, and impacts to performance reviews, and all that. So I find myself having to carve out some percent of my time to stop doing productive work, and "go do AI to show token use." So my workload hasn't changed (or it's gone up), but I have N% less time to work on it because I have to spend time appeasing the AI gods...
Just have an agent chug on a side-project for you, or set up a CI script to review every pull request or some similarly “helpful” task. That should eat a lot of tokens!
Man. If I had this kind of mandate I could really burn some tokens. Review each new PR and extract 100 topics to debate related to it. Spin up 1000 sub agents, each with a different personality profile system prompt, to debate each point until consensus has been reached. Synthesize the learnings into a limerick. Build a Spotify playlist that pairs with the tone of the debates. Post the limerick and link to playlist on the PR and tag me to notify me that I have a PR to review.
Depends on how much you're asked to burn I guess. Until one point, it's actually helpful. Then there's a point where you can do stuff that's semi helpful but doesn't get in the way. But then I would imagine you reach a point where you have to come up with a token burn strategy and some kind of narrative for your manager that's in line with it. I bet at that last point, it gets taxing.
Probably like eating. Having to eat less to loose weight isn't great. Eating anything you want without worries is great. Having to eat more than you want to gain weight, not great.
Oh man, when MCP was still new and shiny I made an MCP that let the AI choose appropriate theme music for what it was doing and it was an absolute blast, I need to make a more modern one.
Peer Gynt Suite's "In the Hall of the Mountain King" made a prominent appearance, but so did Aqua's "Barbie Girl"
I wonder how long this will still be a thing. More and more companies seem to come to the conclusion that tokenmaxxing is just too expensive for the value it delivers. Will there be others that continue doing it? Will there be companies that advertise "unlimited tokens" as part of job listings? How will they react when employees test the limits of "unlimited" (whether by accident or not)?
One of the reasons they exhaust me, is that it's always "one more prompt" to get a UI correct. It's often just slightly off, but it can take 5-10 mins sometimes to rework something. It has led to me working much longer hours.
I think this is in part because I am one of the software engineers that always liked building products more than writing complex software. So, I am driven by the feeling of creating something. And I want to get the feature perfect and complete. But getting from 95%->100% done can take a long time with UI work for me.
Main blocker is I am using apps like Conductor and have lots of plates spinning at once. But that's on, me and I should try and start completing the last part myself.
Maybe when they get better at making SVGs of pelicans riding bicycles, they'll also get better at making UIs that can be reworked into sensible form without too much effort.
I am happy about all the little side-projects, and ideas it help my realize..
Same, but I really have to fight the urge to just add fun new features to things I work on any time inspiration strikes. I am an appalling 'feature factory' if I don't actively keep myself in check. The cost of just building everything is so low, but the value of those things is also incredibly low, so I'm often just bloating what I build.
There's been a lot of articles and posts about the increasing importance of 'taste' in software built with AI, and I'm finding I know need to look for strategies to find some.
I don't have an employer. But most of the excuses I used to tell myself are simply not believable anymore and that causes pressure leading to overworking myself.
I wonder if anybody has an implicit fear that with LLM you're expected to be a 20x engineering all the time, otherwise you're out. Can also lead to people producing shiny apps that impress others (sometimes for legit reasons) even though they have no idea how anything work. A "ship value" culture will not bother with the inner workings or actual skills.
Individual gains from llm seem much larger than net productivity increases. I think a major source of this discrepency is people creating more work for their coworkers at the speed of slop. Especially the people with no idea.
"I did a Chat output, please fix and review it " is the kind of thing that empowers the people who used to have a minimal productivity, and now lets them to wreck things on an industrial scale.
This is valid in the other direction as well. Principle engineers, CTOs, with legitimately earned authority end up using that authority to 100x their output onto the team as if it was a Godsend unlock.
It's not. There is no one person that has universally good taste. Also, we're not in your head, no matter how much better of a coder or whatever. We're not in your head and it's all terribly painful to navigate.
>"I did a Chat output, please fix and review it " is the kind of thing that empowers the people who used to have a minimal productivity, and now lets them to wreck things on an industrial scale.
AI is not a productivity multiplier. There are diminishing results.
The ones that notice the highest increases of productivity are usually the ones that were unproductive at best and dangerously incompetent at worst.
> Individual gains from llm seem much larger than net productivity increases. I think a major source of this discrepency is people creating more work for their coworkers at the speed of slop. Especially the people with no idea.
Lots of companies (nearly all, I’d wager) of any size were leaving bare-minimum a 2x software development speed increase on the table before LLMs, having nothing whatsoever to do with how fast anyone was typing or thinking up code, and everything to do with how they organized and supported development work, and with your basic ordinary corporate dysfunction.
My company, I’d say it was more like 4x or 5x they could have achieved before LLMs, by fixing processes and reducing how often management steps on their own dicks.
All the people I’m seeing with crazy-high LLM productivity at my company? They’ve been given enormous autonomy to basically go do WTF ever they want, and people are jumping to get them anything they need (and most of what they’re doing is prototyping, for that matter). So right off the bat, if they’re competent, they should see a notable multiplier on productivity even if they weren’t using LLMs. Not that those aren’t helping, too, but if you don’t change processes they’re not all that effective, because the problem wasn’t speed of code-writing (and if you can change processes, you already could have sped up development a lot before LLMs…)
In my place I see currently a governance panel effort mandating around LLM agent skills usage, it is so much shit show that I expect productivity is going to fall to 0.5x pre-agents. But not pre-LLM as autocompletion was really helpful in the trenches. The tool in wrong governing hands and you get sand into cogs thrown.
LLMs drive the unit cost of cognition to zero. Therefore, you will exhaust yourself near-instantly trying to drive differentiated value out of cognitive work.
Non-arbitrable labor is one safe haven: bending steel, drilling wells, running cables, flying drones, etc. Physical agency gets you a premium the clankers can’t (yet?) trespass upon. That’s why guys building data centers are making bank & job-hopping while the SAs administering the computational guts of them are struggling.
A second vector is reputational: either by authority (you’re a regulator) or by taste (you’re a rare/reknown specialist) you make quality attestations about cheaply-produced cognitive artifacts.
The first vector is a big community; the second is not.
Get out of being in a knife fight with the clankers on their own turf, they’ll gut you.
Flying drones is an interesting one, I guess you do have to drive the car out to site and set up the drone. But a lot of drone ops are waypointed, automatic flight. I can see a future in which the only thing the operator does is drive the drone van to site, hit the deploy button as the drone pops out the roof, and wait for it to return. Mission set up already by an LLM prompt back at the office.
Then why are so many others in the thread reporting being swamped with requests to review coworkers' slop? If it's genuinely "cognition" at trivial cost, surely this review would be completely unnecessary?
I echo this entirely, brother. I think a lot of us developers have a lot of ideas that were unrealized, and now we have this opportunity to do it. And any time an LLM is sitting idle, it feels like we're wasting our time. Why aren't we having it built something for us? Currently, I work on about three projects at work at the same time and about four personal projects at the same time. My day just zips by. I'll burn four hours without even thinking about it. It's exhausting but exhilarating. I do wonder if burnouts in the future though.
> No one is really pushing me to increase my workload, but at every moment there is always something ready, done by wankers
I confess that the above variant on the quotation is how I originally read it. And that's just about how I feel now with trying to sort through vibe-coded slop projects that are put forth by (well-meaning, probably good intentioned, not evil) people who represent them as if they're the handcrafted result of one dedicated developer.
> In the past (before LLMs) it was already hard to keep up, but now it feels like there's 10x more things waiting at any given time, and there could be 10x more if everyone just "optimized" and streamlined processes fed the AI even more tasks in parallel faster.
I find LLMs to help me manage the unrealistic workload I have, because at least now it's feasible instead of just getting more work piled on top of me with a never ending backlog (that people actually expect me to thin, not let grow). Add on top of that colleagues that would have death by commitee'd many ideas and now just have to argue against actual MVPs that work instead of ideas (or can be proven to not work and discarded without wasting time on them in some cases), and I don't even hate my job as much!
It's just that to ensure that the technology is not a net negative, I need millions upon millions of tokens every single day (tool runs, adversarial reviews, testing), but once you get that inflection point, alongside needing a good enough model, the floor for which currently I'd say GLM 5.2 on Max reasoning reaches, or use something like SOTA Anthropic/OpenAI models, it becomes a pretty good way of working. That said if you have missing pieces there (e.g. using cheap models that aren't very good), the curve of getting stuff done can go downwards and you'll just end up with a lot of slop - useless docs, bad code and an ever increasing amount of technical debt.
On average, each task that I do, needs about 15 minutes to 2 hours of planning and making the agents explore the codebase and refine the plans first.
Curiously, in my case this leads to less burnout cause I can actually pause and grab a drink, meal or go for a walk, while parallel agents do the work, once I've planned things well enough and have dispatched something that will work for 1-4 hours. I don't have to review their output immediately once they finish but can just batch things.
> but at every moment there is always something ready
Yes, this is to me the primary driver of the extreme AI burnout. In ~30 years in Silicon Valley and many, many startups, the pressure has never been as intense.
Before AI I'd mostly work on one thing at a time (at least within a given hour) and in the evening I wouldn't start a new 6 hour task because it's too long, so tomorrow is another day.
Now, that 6 hour task is more like 30 minutes, so there is intense pressure to just knock it off tonight. And then the next one. And one more. And while the bot is thinking, to have 4 other work streams in parallel so there is never, ever, a break in the day. The human mind is not built for 100% utilization 15 hours a day.
> I've noticed LLMs feed my unhealthy "don't want to take a break and waste time being idle" mindset, and I need to correct it.
Embrace it if you’re like me and feel uncomfortable having an idle mind, embrace it! You’ll get more done and being 120% go go go is impossible over the long run so eventually your body will just say I need a break then once you recover full steam head again on the treadmill
From my experience, there are mainly 3 burnout reasons.
1. Multi-tasking is the top one. I usually have to frequently switch between 3 to 5 agent windows which are on different things. It's extremely exhausting when each round takes a few minutes. Before coding agent era, I believe most developers had chance to spend 2+ hours focusing on one thing. Now coding agents have increased my spectrum on the tech stack, but the bandwidth to do deep work isn't increased.
2. Agents are good at getting things running without crash, but do not guarantee to produce correct code. This is quite different from human experts with fundamental knowledge.
3. I also get frustrated when reviewing piles of AI generated low quality PRs. My attention is a limited resource. I don't waste too much energy on other people's work, but if I don't spend more effort, the entire project is corrupted quickly by reckless AI generated code without human author's careful thoughts and designs. Working with people who have less due diligence in mind is painful, working with them in coding agent era is 10x painful because they produce 10x shit. It's a team culture challenge that cannot be easily enforced.
Agreed. I am working hard to restrict myself to only 1-2 agent workflows at a time. More is untenable, though it’s so easy to fall into the trap of deploying an agent “just for this minor fix.”
The worst is that every agent session is generating so many "btw fix this" side-quests that it is really hard to stay in the task focus. I throw some into todo list manually but still it is exploding by the day. Perfect is enemy of good.
I've started feeling slightly physically ill when I read Opus output for hours straight. This article rings very true for me. I've started complaining about it with my team; at least have a personal style guide in your agent rules that eliminates emdashes, the "it's not X, it's Y"s, the long lists of modifiers before the noun, using the word "land" to mean finish, etc. I hope this is just a phase of adolescent LLMs.
"That's such a clever way to see things! Let's delve into that!"
The bots (all of them) seem to show patterns of overuse of specific phrases, words, and punctuation.
Some of those are the ones you mentioned. Another that I've been seeing lately is overuse of the term "gate", wherein: As a human, I know what a gate is. A gate is a thing that can be open, or that can be closed. It might be locked or unlocked. The path beyond the gate may be passable or impassable or nonexistent. The gate is just a gate, and the presence of the gate doesn't imply whether it is open or closed.
But in bot-speak, a gate only refers to a hard block -- an impassable construct. Like a fence or a wall, or even a lava-filled moat.
But while a lava-filled moat is intended to be impassable, the bot uses "gate" -- a thing that is designed to be passed -- to describe that same kind of obstacle.
That's misuse of the term, I think, based on decades of dealing with gates in reality: Usually when I encounter a gate that is closed, I just open it and walk through.
I do have instructions that tell the bot to avoid that usage of the word and it ignores them sometimes anyway.
But "gate" is just today's problem-word that comes to mind as I write this. Yesterday, it was something different. Tomorrow, it will be something else entirely.
The overall pattern here is that of gratingly-repetitive bullshit-grade jargon that doesn't fit to begin with.
I found Codex to use "gate" in a different sense: As the condition of an if statement. I have a local style rule not to use that. Another really grating thing is to use "X-shaped" for "something vaguely related to X". And using temporal words like "still" and "already" in contexts with no temporal connotation.
One thing I did recently with the bot definitely involved actual smoke tests, though: I was working with real hardware that can blow up in real ways, with the bot doing all of the circuit design work and coding based on my goals while I just distantly commanded the show from On-High and plugged shit into a breadboard.
(The project works well and I consider it to be Good Enough; I might go back and polish it more later. There was no smoke, but there could have been.)
My expectation is that you'd hear a lot more about "gated communities", "gatekeeping" etc. than any of the uses of gates that give warm fuzzies. (As a suffix, it's also associated with scandals; but that probably isn't relevant here.)
I was thinking about gated communities earlier today, in fact. We don't have many of them around here.
But where we do have them: At a given time, the gate might be open or closed; passable, or impassable. The presence of a gate is implicit, but the status of that gate is not known without advance knowledge or direct observation. And even when it is closed (even if it defaults to always being closed), there's generally a cromulent way for a person to get that gate to open and then move beyond it. It is designed to be opened and closed.
Gatekeeping: Sure. I've run across a ton of artificial gatekeepers online in my time. I've bypassed countless scores of them. Those are easy: Just ignore them and keep moving.
These aren't examples of the hard-blocking, impassable lava moats that the bot is fond of using "gate" to describe.
Me too. It feels like I’m taking psychic damage from reading so much of this stuff. Contrary to the theory that it’s “just the contract workers’ Nigerian English,” I think the models are developing an ultra-terse hyper-stylized dialect of their own under RL pressure. They seem to be writing increasingly in _code_, and I don’t mean computer code. The words don’t mean quite what they mean to humans.
My non-English-native-speaker head of development, to whom I report, does 100% of his work using LLMs and doesn't even check if the code compiles, but somehow this isn't my biggest problem with it – it's the botspeak in the PR comments (or answers to my PR comments) that are so clearly not written by him, and the documentation that makes absolutely 0 sense sometimes even if I break it down. Just a word salad of "robust", "maintainable", "smoke test" that amount to absolutely nothing. And the "You're absolutely right, I fixed it" responses (narrator: he didn't fix it).
I used to have a lot of fatigue due to it until I stopped caring.
Over the last few days with fable I've found it at times incomprehensible, terse word salad. It also invents phrases assuming I'll understand (but that could be because it's reusing terms in the codebase I no longer remember).
I've often had to paste its output back in to ask it what it actually means. Weird.
I think the main thing is just fatigue. There's so little variety. Each model has its preferred idiolect which everyone becomes tired of due to ubiquity. That's the worst part. It's like always eating fast food.
I was describing this exact feeling today. I haven't quite been able to put it into words but I do get slightly physically ill. Almost similar to mild trypophobia?
`arc land` is burnt into my brain by Phabricator, so I'm aware that the term predates LLMs, but it still drives me nuts.
It's impossible to undo some of these linguistic wobbles. Even if you could filter out 100% of LLM input, the humans themselves are learning to say "land" at a higher frequency now.
This is one of those things I barely noticed because I tend to read fast and skim. Someone pointed the over use of these terms and now its like hitting a set of spike strips every time I'm reading the output from any given model.
Its like when someone points something out a in picture you never saw and now you cannot "unsee" it ever again.
I had Claude make a world cloud from its responses because I was curious to see how big "honest" would be. It barely showed up, so I asked it to just give me the counts and it responded telling me it was trained not to use the word "honest" much because it makes people distrust responses (in addition to showing me the counts).
I just did this on one .claude directory and >20% of the answers there included some variation of "real", "actual", "exact", "honest", "genuine", "valid", "true". ~15% in that directory contain some variation of "real", "genuine" or "honest". This is excluding thinking tokens, sub-agent output, etc.
I’m currently doing a project with someone who only uses LLMs, and it’s exhausting and mentally draining.
Whenever I give feedback on something, the answer is just “let me tell Claude”. The person has no understanding of how everything works, and most of the code reflects that.
The other day he hardcoded in a demo mode, simply because he didn’t even know how to set up a local environment and set environment variables. I’m confused as to why Claude didn’t even knew this, but it might just be the prompting.
I limit LLM usage myself, and if I do use it, I try to use it on extremely specific tasks. It’s the only way it works for me.
I honestly don’t understand how all these companies are getting away with generating AI code. Even in a small project I quickly fall behind on my understanding of the project.
I remember working with people like this before AI and it was annoying but they struggled with productivity because they didn't understand what they were working on enough to produce good code efficiently, so the problem usually took care of itself.
Now these people can thrive because LLM coding encourages the incurious and punishes the deep thinker.
It punishes the inflexible deep thinker, perhaps. If you can’t figure out how to use an incredibly powerful tool to your advantage, how deeply are you thinking, really?
Exactly. Add it to the fact that the world was never particularly "fair" to deep thinkers, because said deep thinkers are often not prioritizing "production". It already rewarded people who could search for and ship a working solution before they understand it perfectly. The person who can only move after fully understanding the problem was at a disadvantage long before LLMs, almost everywhere outside academia.
I'm the same as you. Very specific tasks. Just now I spent two hours in a conversation with claude code to change 5 lines of code that change a significant semantic.
On the other hand my buddy is spending $10k in tokens a day on agents to build something. He's a very smart guy, former developer so it's not just AI psychosis talking.
Still trying to figure it out. Not that I have $10k to spend.
I am feeling very tired. Since I’ve started working with LLMs my output as a solo dev has easily gone up 20x. I’m closing client projects including ones that previously would have been far too ambitious to take on alone. Long running codebases are getting features that have dragged out for months or been sitting in the planning stages for even longer. And overall quality is way up now with more complete (and honestly better) test coverage.
I’m building personal projects at a prodigious pace. In a role reversal I treat the agents like I’m one of my clients (albeit a more technical one who gives them architectural direction) and they are me. I’m using the apps and tooling they make every day. I’ve cancelled SaaS subs for tools I’ve built myself.
I watch the tool calls and realize I should be better at core command line tools so I have a study plan to catch up (just a little bit a day). I’m revisiting long standing config that I dropped in to vim and tmux way back when I started and didn’t know anything.
I guess in theory I could hold my productivity to previous levels and read more. But it doesn’t feel like that’s possible. It feels like we are in one of those sea changes where the promise is less work, but the reality is increased productivity and expectations (the Industrial Revolution feels like the right parallel to reach for). Increased expectations happen in small ways and large. The agents are so good at polishing data presentations that I always send cleaned up visually impactful reports that would have taken significant time in the past just as a matter of course now.
But, I’m tired. I’ve spent the Fable on subscription window sprinting through as much work as I can before it goes API only. (As an aside, I don’t understand how everyone is using so many tokens. I’m sleeping very little and running as much code as I can through fable and I can barely touch a 20x max plan limit.) I keep telling myself I will slow down when it comes off, now it’s extended to the 12th and my window just reset, a few more days to keep knocking out backlog items. I feel like I have to keep the robots busy overnight so when I wake up I can immediately sit down to review. I give directions to agents on my phone which feels wild to me.
The industrial revolution made life better at mass scale but it certainly didn't make life better for the hand loom weaver who had previously went from working at home by hand to higher output but working hard all day in a factory.
From the outside, this all sounds like my father who worked as a mechanic at a flour mill.
His job was mostly watching the machine do the work and then fixing the machine when it broke down. 90% of the time it ran fine.
Going from working by hand to watching the machine would seem really boring and I guess I can understand the AI hate on here from that perspective.
I have a friend who is a confessed LLM-addict and has told me pretty much everything you mentioned (you might be my friend!), I just have a question I want to ask that I never do to him: Why do you feel the need to churn out so many "personal projects" if they burn you out that much? I think we can all agree on the exhaustiveness of reviewing tons of AI-generated code in our $JOB$ - then why extend this unpleasant situation to your whole day?
RE personal projects specifically: addiction is an interesting angle. It feels more like anxiety though, that this thing is going to be taken away and this will be a brief and glorious window when I had the means and opportunity to make all the things I wanted, but never had time for. It’s also exhausting to have a bunch of stuff that you want to do piling up for years.
Could just be a personal thing though. I had kids shortly after I started freelancing. I’m the primary caregiver for them and my partner has always worked insanity hours so just keeping our lives together is a full time job. Every billable hour had to be squeezed from a stone. Maybe I’m trying to make up for lost time.
So the choice is squeezing every ounce of productivity out of me I didn’t even know I had before, burning myself out faster for likely the same amount of money (I doubt you are earning 20x) and not even having the pleasure of coding but just having to argue with a stupid machine that doesn’t even have the decency to get tired or frustrated, yet always falls short of the mark
OR
just keep coding by hand, thinking things in front of a whiteboard when it gets complicated, burning myself out slowly at a more humane pace as we have done for 50+ years of software engineering.
Why is this even a choice? I mean, serious question, do you people have a little bit of self-respect? I am expecting the excuse of “but my boss expects me to use AI”. It is clear most of you have not experienced true burnout, because it’s going to be a world of pain when it hits. Stop turning yourself into machines. You are not.
You’re right I’m not earning 20x more. I recently nearly doubled my rate for new clients and raised existing up by 50%. That’s not enough to offset the efficiency gains in the long run because I will bill fewer hours per client, but right now I’m making more (as said I’m billing a lot of hours). Isn’t that always how it is though with new tech? The power loom operator wasn’t ever going to make 1000x what they made hand looming. A modest increase is I think all workers can expect. The important caveat being only those that survive the destructive reorganization of the field can expect it.
I’m not sure where you work, but I have to compete for every contract (and then to keep them). Hand coding everything isn’t an option anymore or won’t be for much longer. And I don’t even think it’s just freelancers who will be affected. As a solo dev I feel like I’ve been given super powers, but if I worked at a hundred head software house, I’d be looking left and right to see which 30 of us were going to be left when the dust settles.
Not to trivialize your point. I agree we aren’t machines and we should reject being thought of or treated as one. But the reality of software dev IMO has already moved. All reactions are likely over reactions, but I don’t think the swing back is going to land anywhere near where it was.
The reason I'm getting LLM burnout is from dealing with the obvious neutering and opaque downgrading of all the top models.
Prior to the last 12mos AI companies were hell bent on squeezing out the best results from mediocre models.
But... now that the top models have progressed, those same AI companies have switched their efforts into reducing the computation (cost of a producing a result) as much as possible without being too obvious.
What was an exponential slope in the quality of results over the last 36 months has now nearly flat lined.
Addendum: IMHO results have 'flat lined' not because the models aren't much more capable than a year ago, but because conserving the enormous processing cost (of an over subscribed user base) supersedes the goal of following the user's explicit instructions (e.g. especially if that means more processing cost) to generate the best results.
I feel the same way about consumer AI tools now. Gemini and ChatGPT have been abysmal lately. They can no longer be relied on to do multi-turn searching and thinking.
Before, they could stay in thinking mode for more than 7 minutes. For example, "find a source for this claim" would search, analyze, and self-adjust the query. Nowadays, even if I push for it, I cannot make these tools work for more than 30 seconds before they give generic answers, even in "Pro" mode.
At one point we considered adding artifical delay to responses because irrational users dont trust something that finishes fast, even if its the same quality.
How empirical are your comparisons of new and old outputs?
Smart people have been falling into this trap as long as LLMs have hit production. Supposedly early internal versions of GPT-4 had "sparks of AGI" but the public version was "dumbed down for safety"
Hell, the Opus 4.5 moment was only last November, and that was when agentic coding and most coding CLI tools became truly first class options. That's a wild paradigm shift. Hell, GPT-5 wasn't even out (that's August of last year). Most people were using 4o. Their current offerings are wildly better for coding than 4o was.
I generally don't agree with the original commenter here. I think many of the complaints about model regressions are the result of increased usage and increased scrutiny revealing gaps there were there all the time. I've been more critical than most of the output quality since my initial "wow" moment was pretty early - GPT 3.5 API - and the results then were extremely obviously not production ready. But, keeping that level of scrutiny through my usage, I haven't seen the falloff that people who don't look at the output every time claim to see.
But that's also let me use "agent" stuff longer, I guess? The better you were at knowing what you wanted and how to ask for it, the less of an inflection point that you got from Opus 4.5 or GPT 5.
Some of the highest-time-saved-for-max-ROI agentic problems I've solved to date were in September and October of last year with Claude or Cursor.
I do not have the burnout but I certainly operate similarly to the author. I continue to be unable to establish a workflow where allowing the LLM to generate code that I review is faster than writing the code myself. Literally the only two ways out of this dilemma is to blindly trust what was generated or to generate an uncharacteristically exhaustive suite of unit tests to validate every possible scenario. I just write the business logic myself and have the LLM do a lot of the rest. Boilerplate falls into the latter as well.
> generate an uncharacteristically exhaustive suite of unit tests to validate every possible scenario.
This is what you want. You want comprehensive tests at every level, far more than is reasonable for a human to build or maintain, from unit, functional, to full end to end and beyond. Adversarial testing (both TDD-style "write tests to demonstrate this bug", and posthoc "prove this patch wrong with a new test") is the best way to keep AI on track and make those diffs you have to read clean and easy.
An even better way is to use a more strongly typed language and really lock it down, but you can use testing in any language. I feel like my background in TDD and "TATFT" has been secret sauce when working with AI
I see this get mentioned a lot but I still am skeptical that AI can generate tests we can trust more than any other code we know we cannot trust.
Yes tests are conceptually isolated and that helps, but I've personally seen unit tests get generated that are semantically incorrect - that is, they test the structure of the code (e.g. they can check function output types and values), but they can't know _why_ the unit tests need to be there, so the really really helpful tests never get generated. Not to mention the obvious issues with generated tests only testing is x = x, or needless redundant tests for the same thing, or them essentially testing basic features of the language.
You have to iterate on the tests, review and validate them, just like any other code, and if you generate a whole project's tests all at once the quality is abysmal, of course. I've been using a lot of old school data-driven testing techniques, where the harness is just code I review, and the data itself is e.g. json files and drives the system.
I actually have a public (AGPL) example here: https://github.com/pgdogdev/pgdog/tree/main/integration/sql - pgdog is particularly testable since it is trying for complete transparency, so you have a perfect oracle in hand via base postgresql, but it demonstrates the concept at least.
You:
>>> You want comprehensive tests at every level, far more than is reasonable for a human to build or maintain
Also you:
> You have to iterate on the tests, review and validate them
Yes, "maintain" is not quite the same as "review", but the line is veeery fine. I find it really tiring to review masses of tests that an agent spews out.
Especially because I know what it has a tendency to write irrelevant/vacuous/useless tests. It's insane the amount of times I have told Codex to "write a test that reproduces the reported bug, SEE THE TEST FAIL, then implement a fix", only for it to guess an irrelevant test, not run it to see it fail, and implement a code change that has nothing to do with either the test or the actual bug.
Then this falls into the exact same pit the OP mentioned, either you need to blindly trust that the LLM is generating tests that actually work, or you need extensive test coverage for your tests to ensure that your tests are actually testing.
It turns out that you don't actually need tests for your tests, because the code provides a baseline truth for the tests. You do, at some point, have to be epistemically sound enough to actually look for correctness in either the code, behavior, or tests. We unfortunately haven't fully unlocked completely solipsistic value generation yet.
This is also part of why I like end to end tests that use actual UI flow, so I can watch it go by in slow mode before letting it loose fully automated.
Maybe it's because I haven't had my coffee yet, but I cannot understand what you are saying.
What do you mean by "be epistemically sound enough"?
You are using it as if to say "if your code is grounded in sound abstractions, you'll be fine and tests will therefore generate successfully" but preface that claim with "the code provides a baseline truth for the tests". The latter does not follow from the former, and it also does not lift the burden of responsibility away from the programmer - which is where my doubts on test generation stem from in the first place.
Additionally, what is "completely solipsistic value generation"?
You reference it like a perk in a skill tree, but to my ears "generating completely solipsistic values" seems like a way of describing AGI with a philosophical wording instead of just saying AGI.
Which is why test generation has to be carefully guided as well, and this is something at which I've incidentally been fast. Ultimately it's a constant battle between LLM handholding and doing things yourself.
AI shouldn't write tests. At least not all of them. Definitely not e2e's. The tests should be guardrails to constrain agents. This way, the author of code matters less.
I don't even care about tests being correct as you can still verify them even when tedious. What I care is that, more often than not, the shape of the solution is not fixed. Having unit tests for those can be extremely costly as when the changes happens, you have to change all the tests.
I've been burned by this in my honeymoon period with unit testing (pretty much the reason it ended). These days, I prefer broader scope of testing, especially user-facing part. The users may be other developers or end users. I only do unit testing for tricky algorithms or math formulae.
I want all the layers of the pyramid, eventually, but the top layers matter the most. I can't count the number of times my paranoid "make sure that customers can successfully pay us" end to end test suite has prevented the money faucet from being shut off. I install one perennially at any company I work at and they always pay for themselves surprisingly quickly.
I’ve been involved in B2B (so no payment flow). But it’s basically the same with an handful of integration tests for common workflows. They run fast and mostly serve a canary to ensure that we are not crippling some use cases. When a bug hits us, a test case is added/modified for it.
They’re mostly a reflection of the current requirement of the project.
It has good test coverage, mostly unit tests but also a number of end-to-end tests. I also made the LLM build a benchmark, which you can find at the bottom of the readme. It is obviously slow, but I thought that it is good enough to work. When I tried to write a 1 GiB file, I found that it broke down, and after writing half the file, the speed went to under one megabyte per second. Implementation is 10k+ LoC, and I have no idea what is going on there.
That's interesting because I would feed that benchmark back into the agent and loop over it, to see how much faster you could get it, and agents are really good at that kind of recursive optimization. And I would definitely add at least a simulated 1GiB write test, probably a real one honestly, if I was building something like that.
At least with agent-run tests I care about loop speed a lot, but I care about complete coverage more, so having the odd heavy weight full stack integration test is fine, I think.
You're right. This was just a performance issue, but what if next time it is a corruption bug or a security vulnerability or really anything that can cause real consequences if happened in production? I don't think that LLM systems are inherently bound to have this flaw, but I think that we are pretty far from harnesses and algorithms becoming advanced enough so that the LLM system can kind of continuously evaluate its output and ensure it is good in all aspects.
I don't know about that, Fable is, when properly guided, a better engineer for those things than I am. Narrow breadth, weird priorities, myopic and ivory tower as hell, but superhuman. Maybe that says more about me, or maybe not, but certainly it's caught bugs I would not have, and point it at things like a fuzzer, woo buddy, it has been many years since I broke out valgrind and nailed down a memory leak, but it sure can.
100% this is what I've done. I sucked it up and adapted myself to the tool (agents) by having as many implicit guardrails (static typing, functional, no nulls, great linting) and then layering on explicit guardrails (TDD) on top. I also want my workflow to be portable because I don't really trust the frontier model providers.
It is different though. Basically a lot of what I do has changed over the last 2 years. I totally get that a lot of people won't want to adapt though.
Maybe famous last words, but I'm not buying the hype that the "clankers" will take over. I suspect reality will catch up soon and we'll be left with a set of pretty powerful but still limited tools. I see no evidence to the contrary, just investment hype on one side and sky is falling on the other.
The “clankers” won’t take over, but have you also noticed that most people are talking about their workflows/process instead of their results/outcomes? It’s all about “Is the train still running?” than “Are we getting close to the destination?”.
I've just been carefully reading the code. It is easy to slip into just accepting what comes out to speed things up, but reading the code is important.
I save myself by skimming things like tests, templates, some UI. Anything cosmetic. But I have to read the majority of code that ends up on my back end systems.
And for those that have similar-ish sentiments, what mental defect is had that prevents them from just drinking that sweet tasty kool-aid and just use the slop created. What demented trait is in them that causes everyone to just be a stick in the mud trying to ruin everyone else's good time?
In my personal experience, the ones most enthusiastic about LLM magic are those that can't code, but can now walk away with something functional if not quite the best code. Now that they can produce workable code, it will make everyone better. Yet, they have no idea how maintainable the slop is or if it's slop at all.
I actually dispute this, I read all the code, the core thing people have to give up is not "reading the code" per se, it's giving up on "that's not how I would have done it".
When you see a perfectly clear function or object that just isn't your style, you have to accept it and move on. Where there are concrete concerns, or it's unreadable, demand excellence, but treat it like a coworker, not an IDE.
This all reminds me of the differing experiences people had outsourcing coding in the 2010s when it was still called oDesk. You don’t need to read code, you just need to know that the code works. If something doesn’t show up as a problem it doesn’t need to be fixed, and reading code is the least efficient way to discover problems.
The only time I look at code is when something isn’t right and I ask for a root cause analysis. The LLM will show me some offending code or code for reference or evidence and then I quite often say “well that’s dumb you should do it like this instead” but I never need to actually go into the files. I do sometimes look at a git status or git diff.
Yeah this is how I feel about it. Does it look correct? is it doing something weird? Is it forgetting about some gotcha in our domain that it hasn't been taught about yet? Otherwise, ship it.
> When you see a perfectly clear function or object that just isn't your style
is the critical caveat to “that’s not how I would have done it”. Basically, choose your battles because we all have limited bandwidth. So, it’s not really a perfect binary, but a taste that you personally develop.
It sounds kind of like being stuck working with coworkers who--while not overtly hostile--need constant hand-holding and repeat the same kinds of mistakes every day and can't even be genuinely sorry about it.
Just because we work with computers doesn't mean we don't take, er, social-damage. Or perhaps parasocial damage, in this case.
> Some small part of me has started to dread reading LLM output because I know what I’m going to find. False assumptions and hallucinations. Emphatic, staccato fragments. Excessive emojis :sparkle:. :rocket: It’s not just me—these are real patterns (:barf:).
It takes like 5 seconds to add a quick CLAUDE.md / AGENTS.md with a quick style guide, or even just “EMOJI ARE FORBIDDEN” and I find it makes llm output significantly more tolerable. That and a quick style guide with some banned words and phrases.
That won’t help with the false assumptions though, gotta use old fashioned careful reading and critical thinking the catch those.
FWIW, I've been finding that ChatGPT doesn't use emoji at all when I engage with it like a pair programmer and bounce off design ideas, ask for implementation code, propose refactorings etc.
But when I ask it to do data analysis or modeling, the emoji are all over the place, yes.
(And judging by what I've seen on GitHub over the last year or so, I would never in a million years consider asking an LLM to write a project README or documentation unsupervised.)
This is legitimately the reason I'm looking to leave programming.
I got into programming because the problems of programming were interesting to me. But if the problems go from "figure out why this calculator is off by one in France" to "Get this LLM to stop spamming cutsey emojis", then maybe it's time for a career change.
Giving my "otherside", because the pressure to output more at work is real, but at the same time, out side of work, I love this. I'm able to do way more projects than ever before because a barrier to entry was always the amount of research+time required to start up a pet project.
My latest is, I'm really into fizzy/soda water and wanted my own continuous carbonator. My entire build from water source to tap with an ESP32 controlled pump, pressure, water level, cooling fans.
There were so many areas I made mistakes in my shopping cart and it found it - like Home Brewer likes 8mm lines but water filter systems like 9.5mm. Really optimized the versions from a simple on/off pump w/ float switch to effectively a full on PLC system. So many iterations gained by chatting with "someone more experienced". Once I get the parts I can build and have the software side running in less than an hour.
I got into programming to just build stuff, the coding is just a means to an end, try not to think too hard about the how and think more about the why and what
Those who like having a finished thing. Product people. These people love LLMs.
Those who love the process of building a thing, working through a problem, learning something new. Finishing a project is generally not required. For them LLMs are soul sucking hell.
This is just a purity test disguised in high minded rhetoric. Defining "quality" in practice is both impractical and a matter of opinion. Building something first, and making it better later, is it's own form of quality.
We get it, you don't have a passion for the act or the craft, just the end result, but I'm absolutely sick of hearing it all over this site as if it's a universal truth that some of us just don't recognize yet.
Sorry, some of us have a joy for programming where the how is just as important, if not more so, than the what and the why. No matter how much people proclaim that the how doesn't matter to them, it isn't going to suddenly make it true for others.
+1. I’m in the exact opposite camp, I enjoy programming more than “shipping a product”. But the programming itself, coming up with solutions to tough problems, is the fun part. Shipping a product is a side-effect.
But ultimately you got into this craft to solve a problem. That is how the craft developed. And when you build a very complex elaborate system, it can still have interesting technical challenges, even for a developer with AI. You should shift your technical insight to a higher abstraction level, where the AI cannot help anymore.
What’s interesting to me is reasoning about the problem and its implementation. And that doesn’t stop at any abstraction level. Reasoning in the small is just as important as reasoning in the large. And the issue with LLMs is that their capacity for sound reasoning is limited. They are sloppy on any level. You can’t get them to be thorough and dependable in reasoning, regardless of the abstraction level.
Well I think the reasoning of coding agents on lower levels is good enough for me that I don't have to constantly be involved with it, only occasionally have to dive in and help out.
I don’t think that the logical reasoning ability of LLMs depends on the abstraction level. Their heuristic knowledge differs between levels, but that’s a different thing. My concern is the reasoning capabilities.
It may be good enough to make me more productive, but only because I don’t relent on ensuring that the code is well-reasoned. Indeed, I don’t experience that when I do relent.
Ok, so you're basically saying that the AI-generated code does it's job, but when you actually review it you think the way it does it's job is not as it should, and if you get into that flow, your agentic productivity goes down?
If solving jigsaw puzzles with claude will enable the creation of tools that help the people I want to help (students with disabilities), then I would use it for that, without feeling any guilt at all for doing so.
Why should I regret that? Why should I care about your purity tests?
This is dishonest. Your quip was to imply that I'm reducing what is otherwise a fun activity to an automation, on the basis of a purity test.
> Some people merely enjoy programming more than engineering.
And I haven't said otherwise, so I'm not sure what point youre trying to make. My initial goal was to provide my own viewpoint on how to enjoy the process while taking advantage of modern tools, not to tell people they shouldn't enjoy programming more than engineering.
This isn't a response I expect from people who are here for a productive discussion. I'm sorry that you are sick of hearing this, but I'm not responsible for making sure you only read what you find worthy of your own personal brand of respect. Instead of attacking someone for simply offering their point of view, in what appears to be a quasi-gatekeeping effort, maybe you should look inward and discover what is making you this upset toward a complete stranger.
__I cannot take away the joy you have for programming simply by stating what drives me.__
I'm totally fine, just annoyed by how much this "try not to think too hard about the how and think more about the why and what"[0] is getting brought up every single time someone mentions why they prefer hand coding to vibe coding. At this point, it's being overstated, thus gives off less "this is why I use it" and more "come on, get onboard the hypetrain to dystopia or you're going to get left behind". It's hardly conducive to a "productive discussion" when it's the millionth time as a drive-by comment. What kind of response did you expect?
Look, I don't have a problem with your personal motivation. I just hate seeing it suggested that people should abandon their passion because someone else doesn't share it. There's absolutely nothing wrong with "I enjoy making something useful" just as there's nothing wrong with "I enjoy making something with my hands or figuring out how to make it". My problem is with "your enjoyment of that is invalid because I don't enjoy that, so learn to enjoy this".
0: Not really a statement of your drive, is it? More of a directive or suggestion.
Part of the annoying thing is that if you're working on a product which uses LLMs, at some level you run out of levers to pull in terms of being able to fix things. At best you're stacking hacks on top of hacks to prevent unwanted output, but at the end of the day if the LLM really decides it simply doesn't want to follow your instructions, you can't do much other than resign to adding *IMPORTANT* and hoping the next model fixes it.
The experience is much closer to working with an external API that you don't have control over and which simply doesn't do what the documentation says. Those have always been the most frustrating parts of programming, but at least previously you could reverse engineer the actual implementation to work around bugs. You can't even do that now because the "boundary" randomly change every day.
(And I admit I'm salty that the "I don't give a shit about why the calculator doesn't work in France, I'm just here because they pay me to fix it" people were the ones vindicated by technological progress)
I'm no longer in corporate America, so maybe I'm out of touch a bit, but could you just...not...use an LLM? You can still solve interesting problems on your own if you choose to do so?
It’s not there yet but we’re clearly heading towards a world where the answer is “no, you have no choice”. AI is weaved into business processes. If Ai leaves a comment on your pr, you must resolve it before merging, you’re expected to “get things done” at a particular pace consistent with using ai, regardless of whether what you did is any good.
LLM skew the time estimate tho. Now everybody expect stuff based on LLM work instead of normal human work. I/we can choose to solve problem normally, but the expectations have changed.
Yeah at many places you still can. It’s just so easy to turn your brain off and let the robot do a maybe good enough job that even people who know better are merging slop.
We’ve had 3 production incidents this week that slipped past CI because there’s a whole team that is just shoving out PRs without understanding what’s going out.
A lot is said about context you can feed into the LLM but I do think there is still superior power in human context awareness. That kind of ambient collation and organisation of the whole business and its purpose, all the different work going on and how it all relates to eachother. It happens when you isolate business units a bit too much also.
It's not surprising that if you have a hundred separate, isolated contexts working on the same business, that don't cross-talk and have no ability to subconsciously receive and collate, prioritize the thousands of signals we get from our work environment, that you end up shipping lots of incomplete or incompatible work.
There are plenty of shops where they are requiring people to use LLMs. Not "we require you to produce work at X rate" such that one can't hit the target without an LLM, but actually mandating use of LLMs based on the (unproven) assumption that it will boost productivity.
Did you make the move yourself from software development? I'm considering a move into economics lately but I wouldn't want to leave IT completely i.e. saying bye to a 15y career in software and all the stuff I learnt along the way. Ideally I'd find something in between but I don't know how feasible that is.
Well, it’s maybe 120 credit-hours for an Economics bachelor’s degree if you don’t have any prior college credits or degrees, less if you’ve got prior art to build on (or a second major usually only requires the ECON classes).
Fun fact, if you see in the course catalog that’s listed as Remote/Online, check the scheduled times; if there are some, they’re either mandatory meeting times or in-person testing days; if there’s none, then it’s a fully asynchronous class you can THEORETICALLY complete in parallel to your job, whenever you like.
You could dip your toe in the water very slowly the first term and set calendar reminders for the drop & withdraw deadlines. At worst you don’t like it, at middling you realize you can’t multitask school and work, at best you pass the course. One step closer: wax on, wax off.
I get burnout from frustration when the LLM just can't follow instructions.
Like when I'm trying to get it to create an image, and the first pass is beautiful, but ten different request to modify it, with different phrasing and even example images, produce the same image ten times.
Or when you tell it not to use a cheap hack in AGENTS.md about six different ways and in your prompt, and it still does it again, and again.
It's like arguing with an idiot. And THAT gives me burnout.
Also: I've never once seen an emoji in LLM output. What are people talking about?
Excessive amounts of emojis in generated README.md and sometimes in printed/logging outputs. I don't whether this is still an issue because I have a "Never use emojis" instruction in the context.
is there any evidence that Alec Scollon, the first time blog author responsible for this post, even exists? look up the name. boo this post and the premise behind it.
Define "exists". Are you questioning whether a human typed those characters with human fingers on a keyboard? Or are you questioning whether Alec Scollon is the name on that human's government-issued ID? It's not exactly new or unusual for people to use pseudonyms.
It looks that what you describe is partly a "burnout", partly a "sickness" of always the same LLM tricks and output (including errors). Of course LLMs tend to go back to their initial training and even if you "teach" them right, the attention mechanism make them forget things that are not often used (that's the KV-cache) even if they can be important for you (there is room for improvements here).
That said, your reaction is totally human. I personally get sick of how the LLM writes prose with always the same tricks and formulas (even if you prompt it not to). Humans need variants and novelty, that's why fashion exists. We get fed up with repetition and after seeing too much green shoes, seeing a red one is so relieving :) (quick note: I don't like fashion - I'd advocate diversity and personal styles, not fashion)
But that's also the way you work with AI that might be part of the problem. Personally, I don't review all the code the AI generates. I look at it, and I review only the code that matters. And with time on a given project I review less and less because I trust more the architecture and ability of the AI to follow it. In my settings, the AI gets confined to the existing architecture (that we define together at the beginning of the project), and has to ask for authorization to create new things (that's when I review the more). Hoping this could help to avoid burnout myself...
It's burnt me out too. I'm generating 10x more features and multitasking across 4 disparate projects. My greatest concern is I don't really have a strong connection to the underlying fundamentals anymore. I need to see how the things works to internalize it. Now I just trust that the agent wrote this piece correctly.
The productivity drive and the sheer feature set you can generate in record time makes it easy to forget proper sdlc hygiene.
Knowing how things work, knowing what should be possible and where “there be dragons”, and having a pretty well-developed “sixth sense” for all kinds of things is proving just as valuable with LLM-heavy programming as it did before.
… but I am almost certain I’d never have developed those in the first place if I hadn’t spent 25ish years programming on a bunch of different platforms and setting up servers and networks and all that, without LLMs.
I dunno how you make another “me”, now, while before lots and lots of programmers naturally ended up as someone with skills and knowledge like mine, and those skills seem super useful when writing code with LLMs.
I’ve had it three times already in the last 2 years and it is something real. Dizziness and depressive swings is the first that comes, then comes memory loss and finally trouble with speech.
At this point you should stop risking to burn your central cognitive capacity. Be advised.
Note: myself a passionate Claude code user with multi agent parallel approach to dev. Plus 30 yrs of various oldskool dev experience. My blood and sugar and all tests are all within norm, and I bike and swim regularly. I’m not a major drinker and try to avoid alcohol in general. I count 25 trees outside my window and I’m not on any amphétamine pill such as Adderall.
Well the fact you didn’t even mention it means it’s probably the thing that most men your age start suffering from. Cumulative damage and becoming drained from too many orgasms during your lifetime. Sorry bud to break it to you but those are the classic symptoms depression , dizziness, memory loss and speech problems. You need to research how Chinese traditional medicine reverses this. Basically it means 24 months total abstinence or else you will progressively become more cuckish and mentally weakened
The cost of code production is trending to zero. Therefore, the things that come before it and after it become much more crucial.
On the “pre” side, the specification of the problem becomes much more important. On the “post” side: QA and verification that the change has its desired effect, and no ill effects, also becomes much more important.
Sure, these are the next things to be automated, and people will try, but it’s easier on the backend (testing/verification can be automated) than the front end (the spec will be human-written as long as someone cares = forever for brands that matter), there will always be a need for humans on the specification side.
There's some YOLO approach to it, but now Codex has self-approving as well as Claude Code (auto mode). I implemented the same feature by my own on Pi with models through OpenRouter and found results very stable thus I have (as always) limited confidence it can fly.
So (disclaimer: I'm Jujutsu advocate :)) I do "jj new", tell it what to do and then let it run, and check in back later.
If there are things I'm not comfortable (like creating PRs or pushing to repo) I ask it to create Ruby scripts instead named like "__pr.rb" (double underscore files are in my global gitignore). So I can leave it working and then inspect back and edit manually before I run "ruby __pr.rb".
The only thing that's not yet there is tying multiple tmux Claude/Codex session together, but I'm thinking about creating a small Rust app that communicates with Tmux for a preview (or a Ruby script that communicates with my LogSeq directly and manages nodes there :))
Not to minimise the author’s struggle but it sounds like they have relatively minor qualms about the tone of AI text. There are many people, myself included, questioning if this career is still for them. Giving up on writing code as a hobby. That sounds closer to the definition of burnout to me.
> My job has changed from designing and writing code to designing code, describing the design to an LLM, reviewing code the LLM produces
As a long-time engineering manager, PM and, eventually, product owner my response is, "Congrats! You've just been promoted to management." :-)
As a new manager, your first challenge will be successfully delivering commercial results using only a team of 'differently abled' new grad interns. Don't complain, new managers don't get to pick their first team! To be honest, these guys are more like alien brains raised in a vat with no direct senses. They've only ever experienced a data feed of the internet and, oh yeah, they get near-total amnesia a few times a day (but maybe you can teach them to write notes for themselves). They also have ADHD and are somewhere on the spectrum. But don't worry because what they lack in common sense, experience and intuition is offset by having a sort-of photographic memory and a willingness to grind on a problem 24/7. You should be fine. Good luck, we're all counting you...
So, I've also discovered my limits: 4 terminal tabs with claude. Anything above that and my attention gets shredded and I have to reread whole conversations. That being said, after a day of doing 4 sessions simultaneously like this, my brain is fried in the sense that nothing I do can/will relief the stress. It different from normal stress. I completely zone out in the evening.
I saw myself throughout this blog post. It's a bit difficult to come to terms with because I chose this field because I genuinely enjoy coding and building things.
Even projects that used to be challenging enough to impress people with your skills can now be built in 10 minutes with AI just by describing what you want. It's an incredible shift, but it also changes how I think about the craft and what it means to be a good engineer.
I don’t understand what could possibly need to be made so fast that isn’t totally made up billable hours. Running at top speed long enough to be burned out is either ineffective, or valuable enough that someone else can take over while you sleep.
Even with Fabel and all that I constantly keep having to babysit it and correct it like it’s an adolescent and it gets really old and the amount of code. It produces not all of its great at all. I’m burnt out looking at. It’s poor coating that somehow magically works.
I don't have much success with using the LLM to make changes to a big legacy codebase. Instead, I use the LLM to gripe about things I don't like in the code. Usually, it is a brilliant commiserator.
I've taken a bit longer than I wanted but it will be open sourced soon.
It's a durable orchestration engine that takes in specs/requirements and coordinates agents externally (meaning the engine drives the loop, not an agent) until the work is fully implemented/verified and reviewed.
It's meant to be used with any harness as basically the last step. You plan your work with whatever LLM you use and then hand off implementation to the engine (through an MCP server or other surfaces)
It can use your OpenAI/Anthropic subscriptions or any other provider and you can mix and match models across implementation and review in any way you want with fan out for parallel reviewers and more.
The goal is to produce high quality unsupervised code that matches your requirements and is reviewed throughout the implementation rather than at the end only, so that mistakes don't compound.
https://github.com/gastownhall/gascity is certainly a choice. I enjoyed playing with gas town but it was a little too nondeterministic for production code, I think.
Directionally if what you're doing is straightforward it's an amazing experience to be able to slap in an epic planning document and wake up the next day to it being "done", with a big asterisk that done-ness is directly proportional to how good of a spec and how good of a model you were using.
That being said, these days if you use Fable, slap in an epic planning document, and ask it to run a workflow (be sure to specify that subagents should use, say, Sonnet, or wave goodbye to your wallet), it's almost as good as gastown/gascity but far more predictable.
I think having style guidance in your context is valuable for avoiding this kind of thing. Having to read awful, cliched text all day is even worse than having to read reams of useless code. I have some simple humanizing content in there that specifically calls out the rhetorical devices that AI loves, and it drastically improves the diffs and comments. It also makes the coding performance generally slightly worse, but ergonomics uber alles.
If an LLM was a person, it would be known as a "bullshit artist". I've worked with bullshit artists (people) before, and its exhausting trying to separate fact from fiction. It's a reason to avoid working with such people.
A good coworker will admit not knowing something, or if unsure give their best guess but discuss its limitations and why they might be wrong.
Question: Has anyone experimented with using voice to directly prompt an LLM, without doing speech-to-text? If an LLM can pick up on the skeptical nuances in a person's response, it might be prompted not to be overconfident in its subsequent output.
I surprisingly had good results when I told the LLM to only communicate in ASCII memes. It did a fantastic job of summarizing the situation using relevant memes, and the humor was enough to keep things fresh. As silly as it sounds, it's worth trying when you're in that LLM burnout corner.
I don't mind interacting with LLMs myself and find they increase my productivity a decent amount. I just can't stand dealing with other people's slop.
Getting sent IM responses that are copy pasted LLM nonsense. Getting a massive PR to review that was generated overnight and the author didn't read it first.
So annoying, I'm dealing with a client that just constantly feeds my responses to their question into AI. Which of course just asks more questions and tells me how clever I am. I also know the client isn't reading because in many spots I put "[Client Name] you need to answer this directly as we need to know the actual real business detail" and its ignored or the AI provides a detail I know is made up.
I don't really understand how this isn't a self-inflicted problem? Perhaps it's because I'm not really mandated to use LLMs in a particular way, but I've had great success doing a combination of writing code myself and using smaller but faster models as a sort of "flood fill". The larger models can also be useful when you're implementing something which already exists in similar form in the codebase, because you can just put that code in the context and you'll get something very similar outputted. So the more code you write, the better the LLM can be later on. Codebases should get easier to add to the bigger they get, not harder.
Of course if you're supposed to achieve so much output that it's not possible to do anything but vibe it, fair enough.
"My main project right now is to establish a framework for large-scale, unsupervised code generation in our codebase...
sifting through the unsupervised agent’s (Qwen’s) output"
Ouch. I love my local AI setup with Qwen, but that is a mismatch right there. That model is not the right match for that project. It's like trying to develop a major software solution by just throwing in hundreds of fresh junior programmers and have them spew out random code bits, while what you needed is a good PM, a great architect and a handfull of senior engineers. Might as well pack it in for a year until your model has grown into the ability for those rolls. There is a reason why Opus 4.6 and now Fable dramatically changed the SWE capabilities, and IMHO Qwen is not there yet.
started to dread reading LLM output because I know what I’m going to find. False assumptions and hallucinations. Emphatic, staccato fragments. Excessive emojis .
I do not understand these complaints. Yes, those are the defaults and they're annoying, although the general public seems to like them. But you are not stuck with these. You can just tell the LLM how it should interact with you. If you're using any sort of harness beyond the chat window in a web browser, you can codify these instructions in a rules.md file or similar and have it automatically included in any new chat. It's not any harder than changing the default wallpaper or color scheme on your desktop operating system.
In reverse order, you can just tell the LLM to never use emojis. I don't like emphatic staccato fragments either, so I tell it to eschew the language of marketing and hype and stick to a factual and plain language, or to employ an academic tone. I explicitly instruct mine to ask clarifying questions whenever context is ambiguous and to push back on false assumptions or common misconceptions (by me). Hallucinationsa re the biggest problem of those you mention; it's not easy to totally eliminate them (for the same reason it's not easy to instruct people to not fall for scams or disinformation), but you can considerably reduce them by setting standards for citations.
I have ideas about reducing hallucinations over work material (ie a codebase) but am omitting them here as they are not fully thought out or tested.
What helped was a sleep and work system, oriented around being offline that was inspired by nature and from my earlier days in working in tech while car camping across the national parks.
Basically: the sun wins in terms of how all energy on the earth is structured, and expressed. All manners of cycles of organisms and living systems are in relation to its rise and fall, and even its particular color spectrum phases (whether thats night oriented or day). I call this our real circadian rhythm; it's used to being signaled by the light of the sun and maybe fire for millions of years and it isn't until recent centuries when we started tricking our biology with LEDs and lights. So the solution is simple. Orient yourself around the light of the sun and make sure it's the first and last major light source you see; blue limiting is the most important part BEFORE sunrise and AFTER CUT OFF ALL BLUE LIGHT. On my Mac I use a red light filter (using it now, it's 11:07pm ET and the sun went down about 2.5 hours ago). It's really hard to stay alert and chatting with an LLM when the only light sources are red and you keep them dim at that. Our ancestors would rest when the sun's at its peak (~1:05 pm today) and that's a good time to divide my own day productively as well. With intentional breaks diving the middle of the day with sunlight anchoring it, my nervous system is more relaxed, and by the evening time, it's also ready to transition out of anything blue-light assisted and most intellectual work and problem solving falls into this bucket. It's really hard to explain but it really works so simply. To enjoy the process a little more I made this fun sun clock, check it out at https://sunsignal.app
I once experimented with beeswax candles as my only after-dark light source. This meant no hyper-stimulating screen activities whatsoever, too. TV, phone, video games, browsing the web? Nope, nope, nope, and nope. Just dim, warm light from actual flames.
Cured my lifelong “night owl” “trait” in a couple days. Shockingly effective.
Turned out to be hard to keep up and still, like, exist with other people, and you’d probably need to relax it a little in Winter unless your job lets you work reduced hours to kinda “hibernate” (otherwise when would you do anything that’s not work but requires light or electronics?) but it sure worked.
LLMs poison your mind. The more AI slop you read, the more your mind turns into something like slop. This isn't very different from the idea that the food you eat is what your body is made of.
It sounds far fetched at first, but I think it could be true. There was a time in my life when I read literature constantly and started using those patterns in my writing. The habit fell off quickly after I stopped reading for a while, so there might be hope.
I feel this is coming for me. I now have three people generating code using LLMs who have no experience or real understanding of what they're doing. Nor do they have a strong desire to learn, but how could they anyway? I learnt by screwing up time and time again, but each time I went back and learnt why and next time did it better. It would be bad enough if I was managing 5 different agents myself, but it's managing 5 people who are operating 5 agents.
Speak for yourself! This is my third week on SSRI medication and I am the zombie the CEO wants me to be! /s (well, only half /s)
At work, I ended up doing other chores, getting a lot more involved in projects I wouldn't even care to touch. Turns out it's kinda fun being the source of truth at work. I now have a clear sketch of what the company has done and what we can improve on.
Being able to fill in the gaps at a company that doesn't do much feels like a company within a company. Sure, its not Silicon valley, but it's still fun! And job security is guaranteed if you do a bit more than just play the ticket factory
I don't think I have a "burnout", but LLMs are really exhausting due to amount of pressure they generate. No one is really pushing me to increase my workload, but at every moment there is always something ready, done by my clankers or clankers of other people that I could be unblocking. In the past (before LLMs) it was already hard to keep up, but now it feels like there's 10x more things waiting at any given time, and there could be 10x more if everyone just "optimized" and streamlined processes fed the AI even more tasks in parallel faster. It just being a bottleneck of everything, all the time is tiring...
I am happy about all the little side-projects, and ideas it help my realize, and I enjoy exploring this new world, but I've noticed LLMs feed my unhealthy "don't want to take a break and waste time being idle" mindset, and I need to correct it.
W.r.t. article's main complain - I think the similar thing happened due to factory manufacturing automation. What used to be a varied skillful craft in a shop became standing in a single place of an assembly line doing the exact same thing whole day. LLM took away the more creative and variable part of the work, and left the repetitive QA rubber-stamping. Probably some of the mitigations used back then could be rediscovered today.
> done by my clankers or clankers of other people
I'm getting so many requests to review LLM-generated documents - planning docs, docs intended for end-users, project docs, business plan docs. A team member sent me a zip file with about 30 LLM generated documents in it the other day and asked if I could review them right away. And a lot of it was just repetition and/or stuff that was just out of left field, made-up, hallucinated stuff. They're able to generate this stuff way faster than we can review. It used to be that it would take a significant part of a day for a project manager to come up with a planning doc - now they can generate one in a few minutes and send it out for review. It's just really tiring.
The only way to even start to counter that is to make it a firm company policy that if you use an LLM to hallucinate any documents you absolutely must thoroughly review them yourself before you send them to anybody else, and that you are still responsible for the quality of LLM-generated content.
Getting an LLM to vomit out a bunch of documents and sending them straight to another colleague is absolutely unacceptable behaviour.
This is going to just run up against the insanity that is tokenmaxxing every moment of the day. When people are incentivized (upon pain of firing) to get the LLM to vomit out as much as possible, they're hardly going to stop and ponder if schlepping the slop over the wall is acceptable if the alternative is a pink slip.
Which is going to win?
We employees need to remember that most software projects fail. So the work we produce should have lower value than we give it.
We also need to be motivated to stay in our jobs.
Most developers like their projects and value their work. But the chances are that it's for nothing.
Many developers know they work on bad products (gambling industry, military, surveillance, whatever) and so it's here that they focus on their technologies, tools and frameworks rather than the work they produce.
"Agentic engineering" for example.
Id be curious to see what and how Googlers are doing with their 20% time.
The 20% policy at Google is effectively dead.
Feed it into an AI and ask it to adversarially criticize it, doc for doc, send back 30 responses in a zip folder, wipe hands on pants, return to HN.
Don't know if you are serious, but why become part of the problem?
Why not just review a single document quickly, find an error which invalidates the document, and send it back saying "Policy paper 1 mentions X as being on the business plan for Y, it's not on the plan, please can you fix."
Because they will fix it and send again.
Unless you can write a good-sounding reason why it's on them to review a LLM output before sending it to you, they will outsources this reviewing to you, and it's a lot of reviewing.
Why do your colleagues work when they could at least attempt it first?
Because they'll just paste your remarks in their llm, let it correct the text and send it back to you.
This may actually be a solid way to tackle the bullshit asymmetry problem caused by drive-by LLM sloppers.
> now they can generate one in a few minutes and send it out for review.
I think we will very soon move to a prove to me you've read it protocol and/or introduce speed bumps to slow things down.
Thats gonna be a no from me dog. I don’t expect anyone to read something I didn’t read myself
> I'm getting so many requests to review LLM-generated documents
That's the other nightmare of AI slop. So easy to generate endless content. Who will review?
Just today the boss request I review slides for a presentation. But it's all AI slop, generated from querying tickets and docs and who knows what. It's mostly sort of correct but also plenty misleading and incorrect. So now I have to fact check all this slop which will take hours (even with my AI assistance) and rewrite most of it.
If AI didn't exist, he would've had to do the research to generate the content and it would be 99% correct and I could just give a few notes of feedback in 5 minutes. But with the asymmetric AI workload, he can generate it in 5 minutes and I get to spend 3 hours correcting.
Seems like for such requests it's necessary to get some proof of work: require a meeting where for every artifact they sent you to review, they briefly explain the gist and point out the motivation for creating the artifact.
Yeah I heard a similar thing recently at a presentation. At that point wouldn't it be easier to just send the prompt around?
I got this problem with my own employees, LLM are fine, but lazy slop is not permitted. Current idea is to have a clear "best practice" template for most of the research/specs/problem definition they submit and it reduced the slop to a manageable level. But this might work in a smaller company where the management is reading and is strict about these things.
Write an LLM script to review them. Tell it to find at least three severe issues. Set to auto-reply.
He who brings the slop cannon shall be drowned by slop rain.
Rate of generation/Rate of verification is a proxy for signal to noise ratios, just for work.
That ratio has changed, and verification is the hard part.
Verification is the point of all markets (and a decent part of human civ as well).
And review isn’t cost less - https://en.wikipedia.org/wiki/Ironies_of_Automation
> Rate of generation/Rate of verification is a proxy for signal to noise ratios
Hopefully you mean Rate of verification/Rate of generation.
Yes! it should be:
Verification/generation
Wait, what? I thought everyone agrees that modern models post September 2025 (or whenever Opus or whatever 5.6789 was released) do not hallucinate, make things up, contradict themselves and can review their own output into perfection regardless of task, goal or context???? /s
In general I think from the coding side they're more robust now. However, people generating docs are maybe not as experienced with how to prompt in ways that avoid having the LLM tell you what you want to hear. I think this is still a pitfall that can easily be fallen into. Those of us who are doing LLM-assisted coding for the last couple of years are more aware of this now. Those who are planning/management folks are still kind of susceptible depending on how much experience they've had dealing with LLMs.
What a take with no nuance.
> do not hallucinate
They do, just less. To the degree of being usable, as long as there are guardrails and they're used responsibly. For example, if there's code being output, there should be type checking and compilation, as well as code tests that prove that it works or that it doesn't - seeing how abysmal code coverage is in most of the projects I've seem, for whatever reason people thought that they didn't really need it much. They were wrong.
This also implies you need SOTA models on max reasoning.
> make things up
Same as above. Ideally you'd give them some way to verify their claims, like web search or browsing and referencing docs, Jira tickets etc., basically improve the signal to noise ratio.
> contradict themselves
They do so way less than before, as long as the above is true.
> can review their own output into perfection
They are pretty good at reviewing things, especially if you make them do adversarial review! It will never be perfect, but can be close in quality to human output (e.g. the code they produce, when used properly and with intent, is better than the code I've seen many developers write and ship before LLMs were a thing).
This also more or less scales with how much compute you give them - three parallel review agents will turn one output artifact into something good with higher confidence than two, and definitely better than with no review. There's a cost vs quality balance and it seems that all those xhigh and max reasoning modes are still geared way too much towards cost, instead of quality. So you have to make up for that shortcoming yourself.
> regardless of task, goal or context????
Garbage in, garbage out. I won't be an asshole and say that you're holding it wrong, nor will I say that anyone should listen to the claims marketing AI (absolutely delusional takes, meant to attract investors), but we're slowly getting to a better position in regards to LLMs, year by year.
It's just a shame that the peak of inflated expectations hit while the technology still hasn't fully plateaued and reached whatever its ceiling is.
I probably also shouldn't ignore the fact that some people will not care about any of it and send AI generated slop verbatim and to an outside observer there's no way to easily tell apart the difference between the two, unless you make a technical report contain exact references to where the data is sourced from, for example (and then either verify the references yourself, or make another agent do it).
> I think the similar thing happened due to factory manufacturing automation. What used to be a varied skillful craft in a shop became standing in a single place of an assembly line doing the exact same thing whole day.
I had to think of the factory scenes in Charlie Chaplin's Modern Times. The author's feeling is basically the main idea of the sketches, i.e. humans having to follow the pace of the machines instead of the other way around.
Reverse centaurs are nothing new. Ask any worker movement from the last centuries.
> LLM took away the more creative and variable part of the work, and left the repetitive QA rubber-stamping
“I wanted a machine to do the dishes so I could concentrate on my creative work, and all I got was a machine to do my work so I’m left to wash the dishes.”
> I don't think I have a "burnout", but LLMs are really exhausting due to amount of pressure they generate. No one is really pushing me to increase my workload, but at every moment there is always something ready, done by my clankers or clankers of other people that I could be unblocking.
I see a different type of pressure: I'm at a company that still is requiring everyone use LLMs with token leaderboards, time-spent measurements, and impacts to performance reviews, and all that. So I find myself having to carve out some percent of my time to stop doing productive work, and "go do AI to show token use." So my workload hasn't changed (or it's gone up), but I have N% less time to work on it because I have to spend time appeasing the AI gods...
Just have an agent chug on a side-project for you, or set up a CI script to review every pull request or some similarly “helpful” task. That should eat a lot of tokens!
Man. If I had this kind of mandate I could really burn some tokens. Review each new PR and extract 100 topics to debate related to it. Spin up 1000 sub agents, each with a different personality profile system prompt, to debate each point until consensus has been reached. Synthesize the learnings into a limerick. Build a Spotify playlist that pairs with the tone of the debates. Post the limerick and link to playlist on the PR and tag me to notify me that I have a PR to review.
Depends on how much you're asked to burn I guess. Until one point, it's actually helpful. Then there's a point where you can do stuff that's semi helpful but doesn't get in the way. But then I would imagine you reach a point where you have to come up with a token burn strategy and some kind of narrative for your manager that's in line with it. I bet at that last point, it gets taxing.
Probably like eating. Having to eat less to loose weight isn't great. Eating anything you want without worries is great. Having to eat more than you want to gain weight, not great.
Oh man, when MCP was still new and shiny I made an MCP that let the AI choose appropriate theme music for what it was doing and it was an absolute blast, I need to make a more modern one.
Peer Gynt Suite's "In the Hall of the Mountain King" made a prominent appearance, but so did Aqua's "Barbie Girl"
Also, post the limerick to HN.
> side-projects
Just be careful about any legal implication of doing side-projects during work with work-resources.
Ideally, you should ask the LLM to write that CI script.
I wonder how long this will still be a thing. More and more companies seem to come to the conclusion that tokenmaxxing is just too expensive for the value it delivers. Will there be others that continue doing it? Will there be companies that advertise "unlimited tokens" as part of job listings? How will they react when employees test the limits of "unlimited" (whether by accident or not)?
Probably easy for an outsider to say but companies tracking their workers quantatively like this would have me looking for another job
What work are you doing where an LLM can’t meaningfully contribute to your everyday work
One of the reasons they exhaust me, is that it's always "one more prompt" to get a UI correct. It's often just slightly off, but it can take 5-10 mins sometimes to rework something. It has led to me working much longer hours.
I think this is in part because I am one of the software engineers that always liked building products more than writing complex software. So, I am driven by the feeling of creating something. And I want to get the feature perfect and complete. But getting from 95%->100% done can take a long time with UI work for me.
So I work much longer hours now, unfortunately.
It's a bit paradoxical to use AI to increase productivity, and then feel the need to work longer hours to fully actualize said productivity.
But it's probably a common feeling. I wonder if we'll see an increased number of people burn out in the serious, medical sense.
Perhaps do the last 5% yourself?
You're right, I should!
Main blocker is I am using apps like Conductor and have lots of plates spinning at once. But that's on, me and I should try and start completing the last part myself.
This is what I do as I have learnt after much frustration.
Maybe when they get better at making SVGs of pelicans riding bicycles, they'll also get better at making UIs that can be reworked into sensible form without too much effort.
I am happy about all the little side-projects, and ideas it help my realize..
Same, but I really have to fight the urge to just add fun new features to things I work on any time inspiration strikes. I am an appalling 'feature factory' if I don't actively keep myself in check. The cost of just building everything is so low, but the value of those things is also incredibly low, so I'm often just bloating what I build.
There's been a lot of articles and posts about the increasing importance of 'taste' in software built with AI, and I'm finding I know need to look for strategies to find some.
> No one is really pushing me to increase my workload
Spoken like someone who is not at an org/team that has undergone layoffs and reduced hiring in the last 3 years.
You might be in the minority there - especially when it comes to those who are facing burnout.
I don't have an employer. But most of the excuses I used to tell myself are simply not believable anymore and that causes pressure leading to overworking myself.
I wonder if anybody has an implicit fear that with LLM you're expected to be a 20x engineering all the time, otherwise you're out. Can also lead to people producing shiny apps that impress others (sometimes for legit reasons) even though they have no idea how anything work. A "ship value" culture will not bother with the inner workings or actual skills.
Individual gains from llm seem much larger than net productivity increases. I think a major source of this discrepency is people creating more work for their coworkers at the speed of slop. Especially the people with no idea.
"I did a Chat output, please fix and review it " is the kind of thing that empowers the people who used to have a minimal productivity, and now lets them to wreck things on an industrial scale.
This is valid in the other direction as well. Principle engineers, CTOs, with legitimately earned authority end up using that authority to 100x their output onto the team as if it was a Godsend unlock.
It's not. There is no one person that has universally good taste. Also, we're not in your head, no matter how much better of a coder or whatever. We're not in your head and it's all terribly painful to navigate.
I'm just waiting for Bezos to order all work with human agents to use the same API and guardrails as an LLM.
Isn't it how these fulfilment centers already work that way with all these work manuals. Read AMAZON.md /s
>"I did a Chat output, please fix and review it " is the kind of thing that empowers the people who used to have a minimal productivity, and now lets them to wreck things on an industrial scale.
AI is not a productivity multiplier. There are diminishing results.
The ones that notice the highest increases of productivity are usually the ones that were unproductive at best and dangerously incompetent at worst.
> AI is not a productivity multiplier. There are diminishing results.
Sure it is. Just that some of the values being multiplied are negative.
Could you describe your usual workflows and usage patterns with AI?
> Individual gains from llm seem much larger than net productivity increases. I think a major source of this discrepency is people creating more work for their coworkers at the speed of slop. Especially the people with no idea.
Lots of companies (nearly all, I’d wager) of any size were leaving bare-minimum a 2x software development speed increase on the table before LLMs, having nothing whatsoever to do with how fast anyone was typing or thinking up code, and everything to do with how they organized and supported development work, and with your basic ordinary corporate dysfunction.
My company, I’d say it was more like 4x or 5x they could have achieved before LLMs, by fixing processes and reducing how often management steps on their own dicks.
All the people I’m seeing with crazy-high LLM productivity at my company? They’ve been given enormous autonomy to basically go do WTF ever they want, and people are jumping to get them anything they need (and most of what they’re doing is prototyping, for that matter). So right off the bat, if they’re competent, they should see a notable multiplier on productivity even if they weren’t using LLMs. Not that those aren’t helping, too, but if you don’t change processes they’re not all that effective, because the problem wasn’t speed of code-writing (and if you can change processes, you already could have sped up development a lot before LLMs…)
In my place I see currently a governance panel effort mandating around LLM agent skills usage, it is so much shit show that I expect productivity is going to fall to 0.5x pre-agents. But not pre-LLM as autocompletion was really helpful in the trenches. The tool in wrong governing hands and you get sand into cogs thrown.
LLMs drive the unit cost of cognition to zero. Therefore, you will exhaust yourself near-instantly trying to drive differentiated value out of cognitive work. Non-arbitrable labor is one safe haven: bending steel, drilling wells, running cables, flying drones, etc. Physical agency gets you a premium the clankers can’t (yet?) trespass upon. That’s why guys building data centers are making bank & job-hopping while the SAs administering the computational guts of them are struggling. A second vector is reputational: either by authority (you’re a regulator) or by taste (you’re a rare/reknown specialist) you make quality attestations about cheaply-produced cognitive artifacts. The first vector is a big community; the second is not. Get out of being in a knife fight with the clankers on their own turf, they’ll gut you.
Flying drones is an interesting one, I guess you do have to drive the car out to site and set up the drone. But a lot of drone ops are waypointed, automatic flight. I can see a future in which the only thing the operator does is drive the drone van to site, hit the deploy button as the drone pops out the roof, and wait for it to return. Mission set up already by an LLM prompt back at the office.
I bet drone van is the next thing to be automated.
For legal reasons, a human will still need to be in the self-driving van, so the job description will change to "drone van chaperone".
Or 'drone accident scapegoat'.
> LLMs drive the unit cost of cognition to zero.
Then why are so many others in the thread reporting being swamped with requests to review coworkers' slop? If it's genuinely "cognition" at trivial cost, surely this review would be completely unnecessary?
I’d scope it down further than cognition.
Cost of generation has been reduced, and is highly subsidized currently.
Cost of verification has effectively not changed. I’d say as a rule of thumb: verification is the tough part.
Our brains don’t fare well under constant review pressure. https://en.wikipedia.org/wiki/Ironies_of_Automation
I echo this entirely, brother. I think a lot of us developers have a lot of ideas that were unrealized, and now we have this opportunity to do it. And any time an LLM is sitting idle, it feels like we're wasting our time. Why aren't we having it built something for us? Currently, I work on about three projects at work at the same time and about four personal projects at the same time. My day just zips by. I'll burn four hours without even thinking about it. It's exhausting but exhilarating. I do wonder if burnouts in the future though.
> No one is really pushing me to increase my workload, but at every moment there is always something ready, done by wankers
I confess that the above variant on the quotation is how I originally read it. And that's just about how I feel now with trying to sort through vibe-coded slop projects that are put forth by (well-meaning, probably good intentioned, not evil) people who represent them as if they're the handcrafted result of one dedicated developer.
That ALL sounds horrible, is that really life in tech these days?
Yes. Run away if you can, run far away.
> In the past (before LLMs) it was already hard to keep up, but now it feels like there's 10x more things waiting at any given time, and there could be 10x more if everyone just "optimized" and streamlined processes fed the AI even more tasks in parallel faster.
I find LLMs to help me manage the unrealistic workload I have, because at least now it's feasible instead of just getting more work piled on top of me with a never ending backlog (that people actually expect me to thin, not let grow). Add on top of that colleagues that would have death by commitee'd many ideas and now just have to argue against actual MVPs that work instead of ideas (or can be proven to not work and discarded without wasting time on them in some cases), and I don't even hate my job as much!
It's just that to ensure that the technology is not a net negative, I need millions upon millions of tokens every single day (tool runs, adversarial reviews, testing), but once you get that inflection point, alongside needing a good enough model, the floor for which currently I'd say GLM 5.2 on Max reasoning reaches, or use something like SOTA Anthropic/OpenAI models, it becomes a pretty good way of working. That said if you have missing pieces there (e.g. using cheap models that aren't very good), the curve of getting stuff done can go downwards and you'll just end up with a lot of slop - useless docs, bad code and an ever increasing amount of technical debt.
On average, each task that I do, needs about 15 minutes to 2 hours of planning and making the agents explore the codebase and refine the plans first.
Curiously, in my case this leads to less burnout cause I can actually pause and grab a drink, meal or go for a walk, while parallel agents do the work, once I've planned things well enough and have dispatched something that will work for 1-4 hours. I don't have to review their output immediately once they finish but can just batch things.
> but at every moment there is always something ready
Yes, this is to me the primary driver of the extreme AI burnout. In ~30 years in Silicon Valley and many, many startups, the pressure has never been as intense.
Before AI I'd mostly work on one thing at a time (at least within a given hour) and in the evening I wouldn't start a new 6 hour task because it's too long, so tomorrow is another day.
Now, that 6 hour task is more like 30 minutes, so there is intense pressure to just knock it off tonight. And then the next one. And one more. And while the bot is thinking, to have 4 other work streams in parallel so there is never, ever, a break in the day. The human mind is not built for 100% utilization 15 hours a day.
> I've noticed LLMs feed my unhealthy "don't want to take a break and waste time being idle" mindset, and I need to correct it.
Embrace it if you’re like me and feel uncomfortable having an idle mind, embrace it! You’ll get more done and being 120% go go go is impossible over the long run so eventually your body will just say I need a break then once you recover full steam head again on the treadmill
Yeah, but it's that KIND OF AWESOME?
From my experience, there are mainly 3 burnout reasons. 1. Multi-tasking is the top one. I usually have to frequently switch between 3 to 5 agent windows which are on different things. It's extremely exhausting when each round takes a few minutes. Before coding agent era, I believe most developers had chance to spend 2+ hours focusing on one thing. Now coding agents have increased my spectrum on the tech stack, but the bandwidth to do deep work isn't increased. 2. Agents are good at getting things running without crash, but do not guarantee to produce correct code. This is quite different from human experts with fundamental knowledge. 3. I also get frustrated when reviewing piles of AI generated low quality PRs. My attention is a limited resource. I don't waste too much energy on other people's work, but if I don't spend more effort, the entire project is corrupted quickly by reckless AI generated code without human author's careful thoughts and designs. Working with people who have less due diligence in mind is painful, working with them in coding agent era is 10x painful because they produce 10x shit. It's a team culture challenge that cannot be easily enforced.
Agreed. I am working hard to restrict myself to only 1-2 agent workflows at a time. More is untenable, though it’s so easy to fall into the trap of deploying an agent “just for this minor fix.”
The worst is that every agent session is generating so many "btw fix this" side-quests that it is really hard to stay in the task focus. I throw some into todo list manually but still it is exploding by the day. Perfect is enemy of good.
I've started feeling slightly physically ill when I read Opus output for hours straight. This article rings very true for me. I've started complaining about it with my team; at least have a personal style guide in your agent rules that eliminates emdashes, the "it's not X, it's Y"s, the long lists of modifiers before the noun, using the word "land" to mean finish, etc. I hope this is just a phase of adolescent LLMs.
"That's such a clever way to see things! Let's delve into that!"
The bots (all of them) seem to show patterns of overuse of specific phrases, words, and punctuation.
Some of those are the ones you mentioned. Another that I've been seeing lately is overuse of the term "gate", wherein: As a human, I know what a gate is. A gate is a thing that can be open, or that can be closed. It might be locked or unlocked. The path beyond the gate may be passable or impassable or nonexistent. The gate is just a gate, and the presence of the gate doesn't imply whether it is open or closed.
But in bot-speak, a gate only refers to a hard block -- an impassable construct. Like a fence or a wall, or even a lava-filled moat.
But while a lava-filled moat is intended to be impassable, the bot uses "gate" -- a thing that is designed to be passed -- to describe that same kind of obstacle.
That's misuse of the term, I think, based on decades of dealing with gates in reality: Usually when I encounter a gate that is closed, I just open it and walk through.
I do have instructions that tell the bot to avoid that usage of the word and it ignores them sometimes anyway.
But "gate" is just today's problem-word that comes to mind as I write this. Yesterday, it was something different. Tomorrow, it will be something else entirely.
The overall pattern here is that of gratingly-repetitive bullshit-grade jargon that doesn't fit to begin with.
"And that's the real, no-nonsense truth!"
I found Codex to use "gate" in a different sense: As the condition of an if statement. I have a local style rule not to use that. Another really grating thing is to use "X-shaped" for "something vaguely related to X". And using temporal words like "still" and "already" in contexts with no temporal connotation.
I like the word "botspeak".
Another example of typical botspeak is "smoke test". Why not just say "test"? It feels like a way of downplaying the ability to detect problems.
One thing I did recently with the bot definitely involved actual smoke tests, though: I was working with real hardware that can blow up in real ways, with the bot doing all of the circuit design work and coding based on my goals while I just distantly commanded the show from On-High and plugged shit into a breadboard.
(The project works well and I consider it to be Good Enough; I might go back and polish it more later. There was no smoke, but there could have been.)
This makes more sense if you think about the contexts in which people would talk about gates on the Internet, I dare say.
Oh?
I've been on the Internet for ~35 years. What did I miss?
My expectation is that you'd hear a lot more about "gated communities", "gatekeeping" etc. than any of the uses of gates that give warm fuzzies. (As a suffix, it's also associated with scandals; but that probably isn't relevant here.)
I was thinking about gated communities earlier today, in fact. We don't have many of them around here.
But where we do have them: At a given time, the gate might be open or closed; passable, or impassable. The presence of a gate is implicit, but the status of that gate is not known without advance knowledge or direct observation. And even when it is closed (even if it defaults to always being closed), there's generally a cromulent way for a person to get that gate to open and then move beyond it. It is designed to be opened and closed.
Gatekeeping: Sure. I've run across a ton of artificial gatekeepers online in my time. I've bypassed countless scores of them. Those are easy: Just ignore them and keep moving.
These aren't examples of the hard-blocking, impassable lava moats that the bot is fond of using "gate" to describe.
Me too. It feels like I’m taking psychic damage from reading so much of this stuff. Contrary to the theory that it’s “just the contract workers’ Nigerian English,” I think the models are developing an ultra-terse hyper-stylized dialect of their own under RL pressure. They seem to be writing increasingly in _code_, and I don’t mean computer code. The words don’t mean quite what they mean to humans.
My non-English-native-speaker head of development, to whom I report, does 100% of his work using LLMs and doesn't even check if the code compiles, but somehow this isn't my biggest problem with it – it's the botspeak in the PR comments (or answers to my PR comments) that are so clearly not written by him, and the documentation that makes absolutely 0 sense sometimes even if I break it down. Just a word salad of "robust", "maintainable", "smoke test" that amount to absolutely nothing. And the "You're absolutely right, I fixed it" responses (narrator: he didn't fix it).
I used to have a lot of fatigue due to it until I stopped caring.
Over the last few days with fable I've found it at times incomprehensible, terse word salad. It also invents phrases assuming I'll understand (but that could be because it's reusing terms in the codebase I no longer remember).
I've often had to paste its output back in to ask it what it actually means. Weird.
I think the main thing is just fatigue. There's so little variety. Each model has its preferred idiolect which everyone becomes tired of due to ubiquity. That's the worst part. It's like always eating fast food.
"It's like always eating fast food". That right there
I was describing this exact feeling today. I haven't quite been able to put it into words but I do get slightly physically ill. Almost similar to mild trypophobia?
Key points only.
Anything written for humans should be written by humans.
`arc land` is burnt into my brain by Phabricator, so I'm aware that the term predates LLMs, but it still drives me nuts.
It's impossible to undo some of these linguistic wobbles. Even if you could filter out 100% of LLM input, the humans themselves are learning to say "land" at a higher frequency now.
Voice really matters in writing. If everyone uses Opus to write without editing, then it all sounds the same regardless of who it came from.
I had to tell it never to say "hand-wavey" ever again to me. But I agree, I hate the way LLMs phrase sentences.
It's kind of offputting how much Anthropic models these days keep repeating "real", "genuine" and "honest". They've RL'd that way over the top.
"You're right for pointing this out. Honestly, your comment raises a real concern — they genuinely RL'd that over the top." - Claude
This is one of those things I barely noticed because I tend to read fast and skim. Someone pointed the over use of these terms and now its like hitting a set of spike strips every time I'm reading the output from any given model.
Its like when someone points something out a in picture you never saw and now you cannot "unsee" it ever again.
I had Claude make a world cloud from its responses because I was curious to see how big "honest" would be. It barely showed up, so I asked it to just give me the counts and it responded telling me it was trained not to use the word "honest" much because it makes people distrust responses (in addition to showing me the counts).
I just did this on one .claude directory and >20% of the answers there included some variation of "real", "actual", "exact", "honest", "genuine", "valid", "true". ~15% in that directory contain some variation of "real", "genuine" or "honest". This is excluding thinking tokens, sub-agent output, etc.
I’m currently doing a project with someone who only uses LLMs, and it’s exhausting and mentally draining.
Whenever I give feedback on something, the answer is just “let me tell Claude”. The person has no understanding of how everything works, and most of the code reflects that.
The other day he hardcoded in a demo mode, simply because he didn’t even know how to set up a local environment and set environment variables. I’m confused as to why Claude didn’t even knew this, but it might just be the prompting.
I limit LLM usage myself, and if I do use it, I try to use it on extremely specific tasks. It’s the only way it works for me.
I honestly don’t understand how all these companies are getting away with generating AI code. Even in a small project I quickly fall behind on my understanding of the project.
I remember working with people like this before AI and it was annoying but they struggled with productivity because they didn't understand what they were working on enough to produce good code efficiently, so the problem usually took care of itself.
Now these people can thrive because LLM coding encourages the incurious and punishes the deep thinker.
At least to me, before LLMs, those people could somewhat be trained. You could give them feedback, and they would improve and learn the code.
Now your feedback is just another prompt for them, the code might be slightly better, but the person learned nothing from it.
It punishes the inflexible deep thinker, perhaps. If you can’t figure out how to use an incredibly powerful tool to your advantage, how deeply are you thinking, really?
Exactly. Add it to the fact that the world was never particularly "fair" to deep thinkers, because said deep thinkers are often not prioritizing "production". It already rewarded people who could search for and ship a working solution before they understand it perfectly. The person who can only move after fully understanding the problem was at a disadvantage long before LLMs, almost everywhere outside academia.
I'm the same as you. Very specific tasks. Just now I spent two hours in a conversation with claude code to change 5 lines of code that change a significant semantic.
On the other hand my buddy is spending $10k in tokens a day on agents to build something. He's a very smart guy, former developer so it's not just AI psychosis talking.
Still trying to figure it out. Not that I have $10k to spend.
What on earth is he doing with $2.6M in token burn per year? That's at the level where he is running multiple simultaneous frontier models 24/7.
Oh my god, is that in USD?
> $10k in tokens a day
What on earth is he building?
LLM generated code is obtuse and dense, unnecessarily so
I am feeling very tired. Since I’ve started working with LLMs my output as a solo dev has easily gone up 20x. I’m closing client projects including ones that previously would have been far too ambitious to take on alone. Long running codebases are getting features that have dragged out for months or been sitting in the planning stages for even longer. And overall quality is way up now with more complete (and honestly better) test coverage.
I’m building personal projects at a prodigious pace. In a role reversal I treat the agents like I’m one of my clients (albeit a more technical one who gives them architectural direction) and they are me. I’m using the apps and tooling they make every day. I’ve cancelled SaaS subs for tools I’ve built myself.
I watch the tool calls and realize I should be better at core command line tools so I have a study plan to catch up (just a little bit a day). I’m revisiting long standing config that I dropped in to vim and tmux way back when I started and didn’t know anything.
I guess in theory I could hold my productivity to previous levels and read more. But it doesn’t feel like that’s possible. It feels like we are in one of those sea changes where the promise is less work, but the reality is increased productivity and expectations (the Industrial Revolution feels like the right parallel to reach for). Increased expectations happen in small ways and large. The agents are so good at polishing data presentations that I always send cleaned up visually impactful reports that would have taken significant time in the past just as a matter of course now.
But, I’m tired. I’ve spent the Fable on subscription window sprinting through as much work as I can before it goes API only. (As an aside, I don’t understand how everyone is using so many tokens. I’m sleeping very little and running as much code as I can through fable and I can barely touch a 20x max plan limit.) I keep telling myself I will slow down when it comes off, now it’s extended to the 12th and my window just reset, a few more days to keep knocking out backlog items. I feel like I have to keep the robots busy overnight so when I wake up I can immediately sit down to review. I give directions to agents on my phone which feels wild to me.
The industrial revolution made life better at mass scale but it certainly didn't make life better for the hand loom weaver who had previously went from working at home by hand to higher output but working hard all day in a factory.
From the outside, this all sounds like my father who worked as a mechanic at a flour mill.
His job was mostly watching the machine do the work and then fixing the machine when it broke down. 90% of the time it ran fine.
Going from working by hand to watching the machine would seem really boring and I guess I can understand the AI hate on here from that perspective.
I have a friend who is a confessed LLM-addict and has told me pretty much everything you mentioned (you might be my friend!), I just have a question I want to ask that I never do to him: Why do you feel the need to churn out so many "personal projects" if they burn you out that much? I think we can all agree on the exhaustiveness of reviewing tons of AI-generated code in our $JOB$ - then why extend this unpleasant situation to your whole day?
I doubt it, but maybe :)
RE personal projects specifically: addiction is an interesting angle. It feels more like anxiety though, that this thing is going to be taken away and this will be a brief and glorious window when I had the means and opportunity to make all the things I wanted, but never had time for. It’s also exhausting to have a bunch of stuff that you want to do piling up for years.
Could just be a personal thing though. I had kids shortly after I started freelancing. I’m the primary caregiver for them and my partner has always worked insanity hours so just keeping our lives together is a full time job. Every billable hour had to be squeezed from a stone. Maybe I’m trying to make up for lost time.
Classic signs of addiction.
So the choice is squeezing every ounce of productivity out of me I didn’t even know I had before, burning myself out faster for likely the same amount of money (I doubt you are earning 20x) and not even having the pleasure of coding but just having to argue with a stupid machine that doesn’t even have the decency to get tired or frustrated, yet always falls short of the mark
OR
just keep coding by hand, thinking things in front of a whiteboard when it gets complicated, burning myself out slowly at a more humane pace as we have done for 50+ years of software engineering.
Why is this even a choice? I mean, serious question, do you people have a little bit of self-respect? I am expecting the excuse of “but my boss expects me to use AI”. It is clear most of you have not experienced true burnout, because it’s going to be a world of pain when it hits. Stop turning yourself into machines. You are not.
You’re right I’m not earning 20x more. I recently nearly doubled my rate for new clients and raised existing up by 50%. That’s not enough to offset the efficiency gains in the long run because I will bill fewer hours per client, but right now I’m making more (as said I’m billing a lot of hours). Isn’t that always how it is though with new tech? The power loom operator wasn’t ever going to make 1000x what they made hand looming. A modest increase is I think all workers can expect. The important caveat being only those that survive the destructive reorganization of the field can expect it.
I’m not sure where you work, but I have to compete for every contract (and then to keep them). Hand coding everything isn’t an option anymore or won’t be for much longer. And I don’t even think it’s just freelancers who will be affected. As a solo dev I feel like I’ve been given super powers, but if I worked at a hundred head software house, I’d be looking left and right to see which 30 of us were going to be left when the dust settles.
Not to trivialize your point. I agree we aren’t machines and we should reject being thought of or treated as one. But the reality of software dev IMO has already moved. All reactions are likely over reactions, but I don’t think the swing back is going to land anywhere near where it was.
The reason I'm getting LLM burnout is from dealing with the obvious neutering and opaque downgrading of all the top models.
Prior to the last 12mos AI companies were hell bent on squeezing out the best results from mediocre models.
But... now that the top models have progressed, those same AI companies have switched their efforts into reducing the computation (cost of a producing a result) as much as possible without being too obvious.
What was an exponential slope in the quality of results over the last 36 months has now nearly flat lined.
Addendum: IMHO results have 'flat lined' not because the models aren't much more capable than a year ago, but because conserving the enormous processing cost (of an over subscribed user base) supersedes the goal of following the user's explicit instructions (e.g. especially if that means more processing cost) to generate the best results.
I feel the same way about consumer AI tools now. Gemini and ChatGPT have been abysmal lately. They can no longer be relied on to do multi-turn searching and thinking.
Before, they could stay in thinking mode for more than 7 minutes. For example, "find a source for this claim" would search, analyze, and self-adjust the query. Nowadays, even if I push for it, I cannot make these tools work for more than 30 seconds before they give generic answers, even in "Pro" mode.
At one point we considered adding artifical delay to responses because irrational users dont trust something that finishes fast, even if its the same quality.
How empirical are your comparisons of new and old outputs?
Counterpoint: Reels apparently are still addictive
You sure about that? Maybe it is reality hitting expectations after the initial “holy shit” wears off
Smart people have been falling into this trap as long as LLMs have hit production. Supposedly early internal versions of GPT-4 had "sparks of AGI" but the public version was "dumbed down for safety"
https://www.youtube.com/watch?v=qbIk7-JPB2c
I'd bet more on this personally.
This seems hilariously, extremely revisionist.
Hell, the Opus 4.5 moment was only last November, and that was when agentic coding and most coding CLI tools became truly first class options. That's a wild paradigm shift. Hell, GPT-5 wasn't even out (that's August of last year). Most people were using 4o. Their current offerings are wildly better for coding than 4o was.
I generally don't agree with the original commenter here. I think many of the complaints about model regressions are the result of increased usage and increased scrutiny revealing gaps there were there all the time. I've been more critical than most of the output quality since my initial "wow" moment was pretty early - GPT 3.5 API - and the results then were extremely obviously not production ready. But, keeping that level of scrutiny through my usage, I haven't seen the falloff that people who don't look at the output every time claim to see.
But that's also let me use "agent" stuff longer, I guess? The better you were at knowing what you wanted and how to ask for it, the less of an inflection point that you got from Opus 4.5 or GPT 5.
Some of the highest-time-saved-for-max-ROI agentic problems I've solved to date were in September and October of last year with Claude or Cursor.
Cannot relate, my expectations might just gone up, because when I compare what I was producing with agents a year ago vs now, it's night and day.
I do not have the burnout but I certainly operate similarly to the author. I continue to be unable to establish a workflow where allowing the LLM to generate code that I review is faster than writing the code myself. Literally the only two ways out of this dilemma is to blindly trust what was generated or to generate an uncharacteristically exhaustive suite of unit tests to validate every possible scenario. I just write the business logic myself and have the LLM do a lot of the rest. Boilerplate falls into the latter as well.
> generate an uncharacteristically exhaustive suite of unit tests to validate every possible scenario.
This is what you want. You want comprehensive tests at every level, far more than is reasonable for a human to build or maintain, from unit, functional, to full end to end and beyond. Adversarial testing (both TDD-style "write tests to demonstrate this bug", and posthoc "prove this patch wrong with a new test") is the best way to keep AI on track and make those diffs you have to read clean and easy.
An even better way is to use a more strongly typed language and really lock it down, but you can use testing in any language. I feel like my background in TDD and "TATFT" has been secret sauce when working with AI
I see this get mentioned a lot but I still am skeptical that AI can generate tests we can trust more than any other code we know we cannot trust.
Yes tests are conceptually isolated and that helps, but I've personally seen unit tests get generated that are semantically incorrect - that is, they test the structure of the code (e.g. they can check function output types and values), but they can't know _why_ the unit tests need to be there, so the really really helpful tests never get generated. Not to mention the obvious issues with generated tests only testing is x = x, or needless redundant tests for the same thing, or them essentially testing basic features of the language.
You have to iterate on the tests, review and validate them, just like any other code, and if you generate a whole project's tests all at once the quality is abysmal, of course. I've been using a lot of old school data-driven testing techniques, where the harness is just code I review, and the data itself is e.g. json files and drives the system.
I actually have a public (AGPL) example here: https://github.com/pgdogdev/pgdog/tree/main/integration/sql - pgdog is particularly testable since it is trying for complete transparency, so you have a perfect oracle in hand via base postgresql, but it demonstrates the concept at least.
You: >>> You want comprehensive tests at every level, far more than is reasonable for a human to build or maintain
Also you: > You have to iterate on the tests, review and validate them
Yes, "maintain" is not quite the same as "review", but the line is veeery fine. I find it really tiring to review masses of tests that an agent spews out.
Especially because I know what it has a tendency to write irrelevant/vacuous/useless tests. It's insane the amount of times I have told Codex to "write a test that reproduces the reported bug, SEE THE TEST FAIL, then implement a fix", only for it to guess an irrelevant test, not run it to see it fail, and implement a code change that has nothing to do with either the test or the actual bug.
Then this falls into the exact same pit the OP mentioned, either you need to blindly trust that the LLM is generating tests that actually work, or you need extensive test coverage for your tests to ensure that your tests are actually testing.
It turns out that you don't actually need tests for your tests, because the code provides a baseline truth for the tests. You do, at some point, have to be epistemically sound enough to actually look for correctness in either the code, behavior, or tests. We unfortunately haven't fully unlocked completely solipsistic value generation yet.
This is also part of why I like end to end tests that use actual UI flow, so I can watch it go by in slow mode before letting it loose fully automated.
Maybe it's because I haven't had my coffee yet, but I cannot understand what you are saying.
What do you mean by "be epistemically sound enough"?
You are using it as if to say "if your code is grounded in sound abstractions, you'll be fine and tests will therefore generate successfully" but preface that claim with "the code provides a baseline truth for the tests". The latter does not follow from the former, and it also does not lift the burden of responsibility away from the programmer - which is where my doubts on test generation stem from in the first place.
Additionally, what is "completely solipsistic value generation"?
You reference it like a perk in a skill tree, but to my ears "generating completely solipsistic values" seems like a way of describing AGI with a philosophical wording instead of just saying AGI.
Which is why test generation has to be carefully guided as well, and this is something at which I've incidentally been fast. Ultimately it's a constant battle between LLM handholding and doing things yourself.
AI shouldn't write tests. At least not all of them. Definitely not e2e's. The tests should be guardrails to constrain agents. This way, the author of code matters less.
I don't even care about tests being correct as you can still verify them even when tedious. What I care is that, more often than not, the shape of the solution is not fixed. Having unit tests for those can be extremely costly as when the changes happens, you have to change all the tests.
I've been burned by this in my honeymoon period with unit testing (pretty much the reason it ended). These days, I prefer broader scope of testing, especially user-facing part. The users may be other developers or end users. I only do unit testing for tricky algorithms or math formulae.
I want all the layers of the pyramid, eventually, but the top layers matter the most. I can't count the number of times my paranoid "make sure that customers can successfully pay us" end to end test suite has prevented the money faucet from being shut off. I install one perennially at any company I work at and they always pay for themselves surprisingly quickly.
I’ve been involved in B2B (so no payment flow). But it’s basically the same with an handful of integration tests for common workflows. They run fast and mostly serve a canary to ensure that we are not crippling some use cases. When a bug hits us, a test case is added/modified for it.
They’re mostly a reflection of the current requirement of the project.
I used an LLM to build this
https://github.com/dprkh/eventfs
It has good test coverage, mostly unit tests but also a number of end-to-end tests. I also made the LLM build a benchmark, which you can find at the bottom of the readme. It is obviously slow, but I thought that it is good enough to work. When I tried to write a 1 GiB file, I found that it broke down, and after writing half the file, the speed went to under one megabyte per second. Implementation is 10k+ LoC, and I have no idea what is going on there.
That's interesting because I would feed that benchmark back into the agent and loop over it, to see how much faster you could get it, and agents are really good at that kind of recursive optimization. And I would definitely add at least a simulated 1GiB write test, probably a real one honestly, if I was building something like that.
At least with agent-run tests I care about loop speed a lot, but I care about complete coverage more, so having the odd heavy weight full stack integration test is fine, I think.
You're right. This was just a performance issue, but what if next time it is a corruption bug or a security vulnerability or really anything that can cause real consequences if happened in production? I don't think that LLM systems are inherently bound to have this flaw, but I think that we are pretty far from harnesses and algorithms becoming advanced enough so that the LLM system can kind of continuously evaluate its output and ensure it is good in all aspects.
I don't know about that, Fable is, when properly guided, a better engineer for those things than I am. Narrow breadth, weird priorities, myopic and ivory tower as hell, but superhuman. Maybe that says more about me, or maybe not, but certainly it's caught bugs I would not have, and point it at things like a fuzzer, woo buddy, it has been many years since I broke out valgrind and nailed down a memory leak, but it sure can.
Yep.
100% this is what I've done. I sucked it up and adapted myself to the tool (agents) by having as many implicit guardrails (static typing, functional, no nulls, great linting) and then layering on explicit guardrails (TDD) on top. I also want my workflow to be portable because I don't really trust the frontier model providers.
It is different though. Basically a lot of what I do has changed over the last 2 years. I totally get that a lot of people won't want to adapt though.
> I totally get that a lot of people won't want to adapt though.
Or people don't want to be reverse centaur keeping the clankers happily running. Instead of helping to solve users/consumers problem.
Maybe famous last words, but I'm not buying the hype that the "clankers" will take over. I suspect reality will catch up soon and we'll be left with a set of pretty powerful but still limited tools. I see no evidence to the contrary, just investment hype on one side and sky is falling on the other.
The “clankers” won’t take over, but have you also noticed that most people are talking about their workflows/process instead of their results/outcomes? It’s all about “Is the train still running?” than “Are we getting close to the destination?”.
Yeap. Hundreds of tests for small tools are completely trivial now.
I've just been carefully reading the code. It is easy to slip into just accepting what comes out to speed things up, but reading the code is important.
I save myself by skimming things like tests, templates, some UI. Anything cosmetic. But I have to read the majority of code that ends up on my back end systems.
And for those that have similar-ish sentiments, what mental defect is had that prevents them from just drinking that sweet tasty kool-aid and just use the slop created. What demented trait is in them that causes everyone to just be a stick in the mud trying to ruin everyone else's good time?
In my personal experience, the ones most enthusiastic about LLM magic are those that can't code, but can now walk away with something functional if not quite the best code. Now that they can produce workable code, it will make everyone better. Yet, they have no idea how maintainable the slop is or if it's slop at all.
If you can't review the code faster than you could write it yourself, write it yourself.
Don’t read the code!!
I actually dispute this, I read all the code, the core thing people have to give up is not "reading the code" per se, it's giving up on "that's not how I would have done it".
When you see a perfectly clear function or object that just isn't your style, you have to accept it and move on. Where there are concrete concerns, or it's unreadable, demand excellence, but treat it like a coworker, not an IDE.
This all reminds me of the differing experiences people had outsourcing coding in the 2010s when it was still called oDesk. You don’t need to read code, you just need to know that the code works. If something doesn’t show up as a problem it doesn’t need to be fixed, and reading code is the least efficient way to discover problems.
The only time I look at code is when something isn’t right and I ask for a root cause analysis. The LLM will show me some offending code or code for reference or evidence and then I quite often say “well that’s dumb you should do it like this instead” but I never need to actually go into the files. I do sometimes look at a git status or git diff.
Yeah this is how I feel about it. Does it look correct? is it doing something weird? Is it forgetting about some gotcha in our domain that it hasn't been taught about yet? Otherwise, ship it.
> When you see a perfectly clear function or object that just isn't your style
is the critical caveat to “that’s not how I would have done it”. Basically, choose your battles because we all have limited bandwidth. So, it’s not really a perfect binary, but a taste that you personally develop.
"If I want to know something, I’ll probably ask ChatGPT or read Gemini’s overview unless I know what sites I want to check."
This is me. I hate the internet (and search) so much these days that I embrace anything that allows me to not Google a thing.
It sounds kind of like being stuck working with coworkers who--while not overtly hostile--need constant hand-holding and repeat the same kinds of mistakes every day and can't even be genuinely sorry about it.
Just because we work with computers doesn't mean we don't take, er, social-damage. Or perhaps parasocial damage, in this case.
> Some small part of me has started to dread reading LLM output because I know what I’m going to find. False assumptions and hallucinations. Emphatic, staccato fragments. Excessive emojis :sparkle:. :rocket: It’s not just me—these are real patterns (:barf:).
It takes like 5 seconds to add a quick CLAUDE.md / AGENTS.md with a quick style guide, or even just “EMOJI ARE FORBIDDEN” and I find it makes llm output significantly more tolerable. That and a quick style guide with some banned words and phrases.
That won’t help with the false assumptions though, gotta use old fashioned careful reading and critical thinking the catch those.
FWIW, I've been finding that ChatGPT doesn't use emoji at all when I engage with it like a pair programmer and bounce off design ideas, ask for implementation code, propose refactorings etc.
But when I ask it to do data analysis or modeling, the emoji are all over the place, yes.
(And judging by what I've seen on GitHub over the last year or so, I would never in a million years consider asking an LLM to write a project README or documentation unsupervised.)
You are completely right! Do you want me to write these markdown files for you?
</irony>
This is legitimately the reason I'm looking to leave programming.
I got into programming because the problems of programming were interesting to me. But if the problems go from "figure out why this calculator is off by one in France" to "Get this LLM to stop spamming cutsey emojis", then maybe it's time for a career change.
Giving my "otherside", because the pressure to output more at work is real, but at the same time, out side of work, I love this. I'm able to do way more projects than ever before because a barrier to entry was always the amount of research+time required to start up a pet project.
My latest is, I'm really into fizzy/soda water and wanted my own continuous carbonator. My entire build from water source to tap with an ESP32 controlled pump, pressure, water level, cooling fans.
There were so many areas I made mistakes in my shopping cart and it found it - like Home Brewer likes 8mm lines but water filter systems like 9.5mm. Really optimized the versions from a simple on/off pump w/ float switch to effectively a full on PLC system. So many iterations gained by chatting with "someone more experienced". Once I get the parts I can build and have the software side running in less than an hour.
It doesn't make money, but man I really enjoy it.
I got into programming to just build stuff, the coding is just a means to an end, try not to think too hard about the how and think more about the why and what
I got into programming because I like programming, and computers. Not whatever the hell this is.
> Not whatever the hell this is.
do you mean my enjoyment from building things? I'm genuinely confused by this response.
Yes, there are generally two kinds of people:
Those who like having a finished thing. Product people. These people love LLMs.
Those who love the process of building a thing, working through a problem, learning something new. Finishing a project is generally not required. For them LLMs are soul sucking hell.
Yeah, the people that recognize progress is through effort and mastery, and those that would gladly do away with effort, quality be damned.
This is just a purity test disguised in high minded rhetoric. Defining "quality" in practice is both impractical and a matter of opinion. Building something first, and making it better later, is it's own form of quality.
We get it, you don't have a passion for the act or the craft, just the end result, but I'm absolutely sick of hearing it all over this site as if it's a universal truth that some of us just don't recognize yet.
Sorry, some of us have a joy for programming where the how is just as important, if not more so, than the what and the why. No matter how much people proclaim that the how doesn't matter to them, it isn't going to suddenly make it true for others.
+1. I’m in the exact opposite camp, I enjoy programming more than “shipping a product”. But the programming itself, coming up with solutions to tough problems, is the fun part. Shipping a product is a side-effect.
But ultimately you got into this craft to solve a problem. That is how the craft developed. And when you build a very complex elaborate system, it can still have interesting technical challenges, even for a developer with AI. You should shift your technical insight to a higher abstraction level, where the AI cannot help anymore.
What’s interesting to me is reasoning about the problem and its implementation. And that doesn’t stop at any abstraction level. Reasoning in the small is just as important as reasoning in the large. And the issue with LLMs is that their capacity for sound reasoning is limited. They are sloppy on any level. You can’t get them to be thorough and dependable in reasoning, regardless of the abstraction level.
Well I think the reasoning of coding agents on lower levels is good enough for me that I don't have to constantly be involved with it, only occasionally have to dive in and help out.
I don’t think that the logical reasoning ability of LLMs depends on the abstraction level. Their heuristic knowledge differs between levels, but that’s a different thing. My concern is the reasoning capabilities.
And so you experience that AI generated code, even on lower levels, is not good enough for you to be more productive?
It may be good enough to make me more productive, but only because I don’t relent on ensuring that the code is well-reasoned. Indeed, I don’t experience that when I do relent.
Ok, so you're basically saying that the AI-generated code does it's job, but when you actually review it you think the way it does it's job is not as it should, and if you get into that flow, your agentic productivity goes down?
Claude: Solve this jigsaw puzzle for me...
If solving jigsaw puzzles with claude will enable the creation of tools that help the people I want to help (students with disabilities), then I would use it for that, without feeling any guilt at all for doing so.
Why should I regret that? Why should I care about your purity tests?
Nobody said you should? Some people merely enjoy programming more than engineering.
> Nobody said you should?
This is dishonest. Your quip was to imply that I'm reducing what is otherwise a fun activity to an automation, on the basis of a purity test.
> Some people merely enjoy programming more than engineering.
And I haven't said otherwise, so I'm not sure what point youre trying to make. My initial goal was to provide my own viewpoint on how to enjoy the process while taking advantage of modern tools, not to tell people they shouldn't enjoy programming more than engineering.
You okay?
This isn't a response I expect from people who are here for a productive discussion. I'm sorry that you are sick of hearing this, but I'm not responsible for making sure you only read what you find worthy of your own personal brand of respect. Instead of attacking someone for simply offering their point of view, in what appears to be a quasi-gatekeeping effort, maybe you should look inward and discover what is making you this upset toward a complete stranger.
__I cannot take away the joy you have for programming simply by stating what drives me.__
I'm totally fine, just annoyed by how much this "try not to think too hard about the how and think more about the why and what"[0] is getting brought up every single time someone mentions why they prefer hand coding to vibe coding. At this point, it's being overstated, thus gives off less "this is why I use it" and more "come on, get onboard the hypetrain to dystopia or you're going to get left behind". It's hardly conducive to a "productive discussion" when it's the millionth time as a drive-by comment. What kind of response did you expect?
Look, I don't have a problem with your personal motivation. I just hate seeing it suggested that people should abandon their passion because someone else doesn't share it. There's absolutely nothing wrong with "I enjoy making something useful" just as there's nothing wrong with "I enjoy making something with my hands or figuring out how to make it". My problem is with "your enjoyment of that is invalid because I don't enjoy that, so learn to enjoy this".
0: Not really a statement of your drive, is it? More of a directive or suggestion.
> At this point, it's being overstated
then so is your opinion on the matter. this goes both ways. I can just as easily dismiss your commentary as "drive-by".
> Not really a statement of your drive, is it? More of a directive or suggestion.
Yeah, so was my initial comment. Merely a suggestion concerning view points, and how a shift in that can bring back some amount of joy.
You don't have the moral high ground here.
Part of the annoying thing is that if you're working on a product which uses LLMs, at some level you run out of levers to pull in terms of being able to fix things. At best you're stacking hacks on top of hacks to prevent unwanted output, but at the end of the day if the LLM really decides it simply doesn't want to follow your instructions, you can't do much other than resign to adding *IMPORTANT* and hoping the next model fixes it.
The experience is much closer to working with an external API that you don't have control over and which simply doesn't do what the documentation says. Those have always been the most frustrating parts of programming, but at least previously you could reverse engineer the actual implementation to work around bugs. You can't even do that now because the "boundary" randomly change every day.
(And I admit I'm salty that the "I don't give a shit about why the calculator doesn't work in France, I'm just here because they pay me to fix it" people were the ones vindicated by technological progress)
I'm no longer in corporate America, so maybe I'm out of touch a bit, but could you just...not...use an LLM? You can still solve interesting problems on your own if you choose to do so?
It’s not there yet but we’re clearly heading towards a world where the answer is “no, you have no choice”. AI is weaved into business processes. If Ai leaves a comment on your pr, you must resolve it before merging, you’re expected to “get things done” at a particular pace consistent with using ai, regardless of whether what you did is any good.
LLM skew the time estimate tho. Now everybody expect stuff based on LLM work instead of normal human work. I/we can choose to solve problem normally, but the expectations have changed.
Yeah at many places you still can. It’s just so easy to turn your brain off and let the robot do a maybe good enough job that even people who know better are merging slop.
We’ve had 3 production incidents this week that slipped past CI because there’s a whole team that is just shoving out PRs without understanding what’s going out.
A lot is said about context you can feed into the LLM but I do think there is still superior power in human context awareness. That kind of ambient collation and organisation of the whole business and its purpose, all the different work going on and how it all relates to eachother. It happens when you isolate business units a bit too much also.
It's not surprising that if you have a hundred separate, isolated contexts working on the same business, that don't cross-talk and have no ability to subconsciously receive and collate, prioritize the thousands of signals we get from our work environment, that you end up shipping lots of incomplete or incompatible work.
There are plenty of shops where they are requiring people to use LLMs. Not "we require you to produce work at X rate" such that one can't hit the target without an LLM, but actually mandating use of LLMs based on the (unproven) assumption that it will boost productivity.
Accounting is desperate for accountants because they’re necessary for legal and compliance reasons. Join up today!
Did you make the move yourself from software development? I'm considering a move into economics lately but I wouldn't want to leave IT completely i.e. saying bye to a 15y career in software and all the stuff I learnt along the way. Ideally I'd find something in between but I don't know how feasible that is.
Well, it’s maybe 120 credit-hours for an Economics bachelor’s degree if you don’t have any prior college credits or degrees, less if you’ve got prior art to build on (or a second major usually only requires the ECON classes).
Fun fact, if you see in the course catalog that’s listed as Remote/Online, check the scheduled times; if there are some, they’re either mandatory meeting times or in-person testing days; if there’s none, then it’s a fully asynchronous class you can THEORETICALLY complete in parallel to your job, whenever you like.
You could dip your toe in the water very slowly the first term and set calendar reminders for the drop & withdraw deadlines. At worst you don’t like it, at middling you realize you can’t multitask school and work, at best you pass the course. One step closer: wax on, wax off.
Doesn't this require accreditation that most programmers wouldn't have?
6 terms left :D
Leave programing for what though?
That is the conundrum, and I genuinely enjoy building software.
Anything else, dude. Let’s do without the propaganda that programming is a cushy job, and any other job is menial underpaid grunt work.
iwould advise that if you love it, stay and coast.
this AI bubble will pop. when it does you'll be hot stuff all over again.
I get burnout from frustration when the LLM just can't follow instructions.
Like when I'm trying to get it to create an image, and the first pass is beautiful, but ten different request to modify it, with different phrasing and even example images, produce the same image ten times. Or when you tell it not to use a cheap hack in AGENTS.md about six different ways and in your prompt, and it still does it again, and again.
It's like arguing with an idiot. And THAT gives me burnout.
Also: I've never once seen an emoji in LLM output. What are people talking about?
> Also: I've never once seen an emoji in LLM output. What are people talking about?
It seems to be heavily dependent on the task.
Excessive amounts of emojis in generated README.md and sometimes in printed/logging outputs. I don't whether this is still an issue because I have a "Never use emojis" instruction in the context.
is there any evidence that Alec Scollon, the first time blog author responsible for this post, even exists? look up the name. boo this post and the premise behind it.
Define "exists". Are you questioning whether a human typed those characters with human fingers on a keyboard? Or are you questioning whether Alec Scollon is the name on that human's government-issued ID? It's not exactly new or unusual for people to use pseudonyms.
The domain "alecscollon.com" was also registered today, according to WHOIS
I don't even know what percentage of people here are even real. I don't like any of this any more.
It looks that what you describe is partly a "burnout", partly a "sickness" of always the same LLM tricks and output (including errors). Of course LLMs tend to go back to their initial training and even if you "teach" them right, the attention mechanism make them forget things that are not often used (that's the KV-cache) even if they can be important for you (there is room for improvements here).
That said, your reaction is totally human. I personally get sick of how the LLM writes prose with always the same tricks and formulas (even if you prompt it not to). Humans need variants and novelty, that's why fashion exists. We get fed up with repetition and after seeing too much green shoes, seeing a red one is so relieving :) (quick note: I don't like fashion - I'd advocate diversity and personal styles, not fashion)
But that's also the way you work with AI that might be part of the problem. Personally, I don't review all the code the AI generates. I look at it, and I review only the code that matters. And with time on a given project I review less and less because I trust more the architecture and ability of the AI to follow it. In my settings, the AI gets confined to the existing architecture (that we define together at the beginning of the project), and has to ask for authorization to create new things (that's when I review the more). Hoping this could help to avoid burnout myself...
It's burnt me out too. I'm generating 10x more features and multitasking across 4 disparate projects. My greatest concern is I don't really have a strong connection to the underlying fundamentals anymore. I need to see how the things works to internalize it. Now I just trust that the agent wrote this piece correctly.
The productivity drive and the sheer feature set you can generate in record time makes it easy to forget proper sdlc hygiene.
Knowing how things work, knowing what should be possible and where “there be dragons”, and having a pretty well-developed “sixth sense” for all kinds of things is proving just as valuable with LLM-heavy programming as it did before.
… but I am almost certain I’d never have developed those in the first place if I hadn’t spent 25ish years programming on a bunch of different platforms and setting up servers and networks and all that, without LLMs.
I dunno how you make another “me”, now, while before lots and lots of programmers naturally ended up as someone with skills and knowledge like mine, and those skills seem super useful when writing code with LLMs.
I’ve had it three times already in the last 2 years and it is something real. Dizziness and depressive swings is the first that comes, then comes memory loss and finally trouble with speech.
At this point you should stop risking to burn your central cognitive capacity. Be advised.
Note: myself a passionate Claude code user with multi agent parallel approach to dev. Plus 30 yrs of various oldskool dev experience. My blood and sugar and all tests are all within norm, and I bike and swim regularly. I’m not a major drinker and try to avoid alcohol in general. I count 25 trees outside my window and I’m not on any amphétamine pill such as Adderall.
Well the fact you didn’t even mention it means it’s probably the thing that most men your age start suffering from. Cumulative damage and becoming drained from too many orgasms during your lifetime. Sorry bud to break it to you but those are the classic symptoms depression , dizziness, memory loss and speech problems. You need to research how Chinese traditional medicine reverses this. Basically it means 24 months total abstinence or else you will progressively become more cuckish and mentally weakened
I can't remember the last time I commented on HN, but: what the fuck?
The cost of code production is trending to zero. Therefore, the things that come before it and after it become much more crucial.
On the “pre” side, the specification of the problem becomes much more important. On the “post” side: QA and verification that the change has its desired effect, and no ill effects, also becomes much more important.
Sure, these are the next things to be automated, and people will try, but it’s easier on the backend (testing/verification can be automated) than the front end (the spec will be human-written as long as someone cares = forever for brands that matter), there will always be a need for humans on the specification side.
I have found (partial) remedy: let it go.
There's some YOLO approach to it, but now Codex has self-approving as well as Claude Code (auto mode). I implemented the same feature by my own on Pi with models through OpenRouter and found results very stable thus I have (as always) limited confidence it can fly.
So (disclaimer: I'm Jujutsu advocate :)) I do "jj new", tell it what to do and then let it run, and check in back later.
If there are things I'm not comfortable (like creating PRs or pushing to repo) I ask it to create Ruby scripts instead named like "__pr.rb" (double underscore files are in my global gitignore). So I can leave it working and then inspect back and edit manually before I run "ruby __pr.rb".
The only thing that's not yet there is tying multiple tmux Claude/Codex session together, but I'm thinking about creating a small Rust app that communicates with Tmux for a preview (or a Ruby script that communicates with my LogSeq directly and manages nodes there :))
I’m experimenting with the following in my settings:
“Look at the first letter of the prompt, and use the style of the matching literary stylist enumerated here:
For English:
A — W. H. Auden
B — Bill Bryson
C — Italo Calvino
D — Joan Didion
E — T. S. Eliot
F — William Faulkner
G — Gabriel García Márquez
…”
I certainly don’t see emojis any more.
Not to minimise the author’s struggle but it sounds like they have relatively minor qualms about the tone of AI text. There are many people, myself included, questioning if this career is still for them. Giving up on writing code as a hobby. That sounds closer to the definition of burnout to me.
If you can afford it, I highly recommend quitting and going independent at least for a while. It’s so much fun building right now.
Can you elaborate on what you're doing to make a living?
I'm a SWE and still a SWE. Just moved from HCOL to LCOL and am working for myself.
> My job has changed from designing and writing code to designing code, describing the design to an LLM, reviewing code the LLM produces
As a long-time engineering manager, PM and, eventually, product owner my response is, "Congrats! You've just been promoted to management." :-)
As a new manager, your first challenge will be successfully delivering commercial results using only a team of 'differently abled' new grad interns. Don't complain, new managers don't get to pick their first team! To be honest, these guys are more like alien brains raised in a vat with no direct senses. They've only ever experienced a data feed of the internet and, oh yeah, they get near-total amnesia a few times a day (but maybe you can teach them to write notes for themselves). They also have ADHD and are somewhere on the spectrum. But don't worry because what they lack in common sense, experience and intuition is offset by having a sort-of photographic memory and a willingness to grind on a problem 24/7. You should be fine. Good luck, we're all counting you...
As a former manager, I am sorry, but the last paragraph just sounds like normal rookie team members.
Including myself as well. It's how you grow into the role.
So, I've also discovered my limits: 4 terminal tabs with claude. Anything above that and my attention gets shredded and I have to reread whole conversations. That being said, after a day of doing 4 sessions simultaneously like this, my brain is fried in the sense that nothing I do can/will relief the stress. It different from normal stress. I completely zone out in the evening.
I saw myself throughout this blog post. It's a bit difficult to come to terms with because I chose this field because I genuinely enjoy coding and building things.
Even projects that used to be challenging enough to impress people with your skills can now be built in 10 minutes with AI just by describing what you want. It's an incredible shift, but it also changes how I think about the craft and what it means to be a good engineer.
LLM burnout is when you catch yourself asking ChatGPT 'how are you' and genuinely waiting for an answer that isn't 'I'm an AI language model'
I don’t understand what could possibly need to be made so fast that isn’t totally made up billable hours. Running at top speed long enough to be burned out is either ineffective, or valuable enough that someone else can take over while you sleep.
Even with Fabel and all that I constantly keep having to babysit it and correct it like it’s an adolescent and it gets really old and the amount of code. It produces not all of its great at all. I’m burnt out looking at. It’s poor coating that somehow magically works.
I don't have much success with using the LLM to make changes to a big legacy codebase. Instead, I use the LLM to gripe about things I don't like in the code. Usually, it is a brilliant commiserator.
What to do to achieve this kind of burn out, I feel like I am a stubborn old developer, I'm coding since 2013 and I am 25 years old.
My mind still can't function well without having knowledge about everything.
> My main project right now is to establish a framework for large-scale, unsupervised code generation in our codebase
Anyone else working on something like this or know of any projects attempting it?
I'm building something for that.
I've taken a bit longer than I wanted but it will be open sourced soon.
It's a durable orchestration engine that takes in specs/requirements and coordinates agents externally (meaning the engine drives the loop, not an agent) until the work is fully implemented/verified and reviewed.
It's meant to be used with any harness as basically the last step. You plan your work with whatever LLM you use and then hand off implementation to the engine (through an MCP server or other surfaces)
It can use your OpenAI/Anthropic subscriptions or any other provider and you can mix and match models across implementation and review in any way you want with fan out for parallel reviewers and more.
The goal is to produce high quality unsupervised code that matches your requirements and is reviewed throughout the implementation rather than at the end only, so that mistakes don't compound.
https://engine.build if you want to get notified when it releases.
https://github.com/gastownhall/gascity is certainly a choice. I enjoyed playing with gas town but it was a little too nondeterministic for production code, I think.
Directionally if what you're doing is straightforward it's an amazing experience to be able to slap in an epic planning document and wake up the next day to it being "done", with a big asterisk that done-ness is directly proportional to how good of a spec and how good of a model you were using.
That being said, these days if you use Fable, slap in an epic planning document, and ask it to run a workflow (be sure to specify that subagents should use, say, Sonnet, or wave goodbye to your wallet), it's almost as good as gastown/gascity but far more predictable.
An almost infinite supply of such modern-day alchemists.
I think having style guidance in your context is valuable for avoiding this kind of thing. Having to read awful, cliched text all day is even worse than having to read reams of useless code. I have some simple humanizing content in there that specifically calls out the rhetorical devices that AI loves, and it drastically improves the diffs and comments. It also makes the coding performance generally slightly worse, but ergonomics uber alles.
If an LLM was a person, it would be known as a "bullshit artist". I've worked with bullshit artists (people) before, and its exhausting trying to separate fact from fiction. It's a reason to avoid working with such people.
A good coworker will admit not knowing something, or if unsure give their best guess but discuss its limitations and why they might be wrong.
Question: Has anyone experimented with using voice to directly prompt an LLM, without doing speech-to-text? If an LLM can pick up on the skeptical nuances in a person's response, it might be prompted not to be overconfident in its subsequent output.
It's amazing how many do not have experience with the full pre and post SDLC.
Tell the llm to answer like a cavemen, if llm talk like cavemen, the answers become shorter and more compressed.
https://github.com/JuliusBrussee/caveman
It's for getting it to output shorter answers, but also could help with your burnout.
I surprisingly had good results when I told the LLM to only communicate in ASCII memes. It did a fantastic job of summarizing the situation using relevant memes, and the humor was enough to keep things fresh. As silly as it sounds, it's worth trying when you're in that LLM burnout corner.
What does that have to do with the topic? Someone is sharing something serious on this thread, I don't see how your message helps.
Works for me.
Burn out? This guy is weakling with zero experience in full QA and validation.
Come on, Claude is not that bad.
Current AI is like the film company producing TV series or movies
Review AI code line by line is like watch movies frame by frame, and is impossible, very difficult, terribly boring, or abandoned sooner or later.
Avoid Gemini and the lesser ChatGPT models and your emoji problem goes away.
Also known as clanker fatigue
I don't mind interacting with LLMs myself and find they increase my productivity a decent amount. I just can't stand dealing with other people's slop.
Getting sent IM responses that are copy pasted LLM nonsense. Getting a massive PR to review that was generated overnight and the author didn't read it first.
So annoying, I'm dealing with a client that just constantly feeds my responses to their question into AI. Which of course just asks more questions and tells me how clever I am. I also know the client isn't reading because in many spots I put "[Client Name] you need to answer this directly as we need to know the actual real business detail" and its ignored or the AI provides a detail I know is made up.
I don't really understand how this isn't a self-inflicted problem? Perhaps it's because I'm not really mandated to use LLMs in a particular way, but I've had great success doing a combination of writing code myself and using smaller but faster models as a sort of "flood fill". The larger models can also be useful when you're implementing something which already exists in similar form in the codebase, because you can just put that code in the context and you'll get something very similar outputted. So the more code you write, the better the LLM can be later on. Codebases should get easier to add to the bigger they get, not harder.
Of course if you're supposed to achieve so much output that it's not possible to do anything but vibe it, fair enough.
"My main project right now is to establish a framework for large-scale, unsupervised code generation in our codebase... sifting through the unsupervised agent’s (Qwen’s) output"
Ouch. I love my local AI setup with Qwen, but that is a mismatch right there. That model is not the right match for that project. It's like trying to develop a major software solution by just throwing in hundreds of fresh junior programmers and have them spew out random code bits, while what you needed is a good PM, a great architect and a handfull of senior engineers. Might as well pack it in for a year until your model has grown into the ability for those rolls. There is a reason why Opus 4.6 and now Fable dramatically changed the SWE capabilities, and IMHO Qwen is not there yet.
i think we're interacting with a character not a person.
I feel you. 'But we need to adapt or quickly become invaluable'. It is a tough reality to swallow - me included.
started to dread reading LLM output because I know what I’m going to find. False assumptions and hallucinations. Emphatic, staccato fragments. Excessive emojis .
I do not understand these complaints. Yes, those are the defaults and they're annoying, although the general public seems to like them. But you are not stuck with these. You can just tell the LLM how it should interact with you. If you're using any sort of harness beyond the chat window in a web browser, you can codify these instructions in a rules.md file or similar and have it automatically included in any new chat. It's not any harder than changing the default wallpaper or color scheme on your desktop operating system.
In reverse order, you can just tell the LLM to never use emojis. I don't like emphatic staccato fragments either, so I tell it to eschew the language of marketing and hype and stick to a factual and plain language, or to employ an academic tone. I explicitly instruct mine to ask clarifying questions whenever context is ambiguous and to push back on false assumptions or common misconceptions (by me). Hallucinationsa re the biggest problem of those you mention; it's not easy to totally eliminate them (for the same reason it's not easy to instruct people to not fall for scams or disinformation), but you can considerably reduce them by setting standards for citations.
I have ideas about reducing hallucinations over work material (ie a codebase) but am omitting them here as they are not fully thought out or tested.
We are all going to discover the Ironies of Automation.
https://en.wikipedia.org/wiki/Ironies_of_Automation
https://ckrybus.com/static/papers/Bainbridge_1983_Automatica...
you are the one accountable here, not the llms
I had this too. Started after lay off in October.
What helped was a sleep and work system, oriented around being offline that was inspired by nature and from my earlier days in working in tech while car camping across the national parks.
Basically: the sun wins in terms of how all energy on the earth is structured, and expressed. All manners of cycles of organisms and living systems are in relation to its rise and fall, and even its particular color spectrum phases (whether thats night oriented or day). I call this our real circadian rhythm; it's used to being signaled by the light of the sun and maybe fire for millions of years and it isn't until recent centuries when we started tricking our biology with LEDs and lights. So the solution is simple. Orient yourself around the light of the sun and make sure it's the first and last major light source you see; blue limiting is the most important part BEFORE sunrise and AFTER CUT OFF ALL BLUE LIGHT. On my Mac I use a red light filter (using it now, it's 11:07pm ET and the sun went down about 2.5 hours ago). It's really hard to stay alert and chatting with an LLM when the only light sources are red and you keep them dim at that. Our ancestors would rest when the sun's at its peak (~1:05 pm today) and that's a good time to divide my own day productively as well. With intentional breaks diving the middle of the day with sunlight anchoring it, my nervous system is more relaxed, and by the evening time, it's also ready to transition out of anything blue-light assisted and most intellectual work and problem solving falls into this bucket. It's really hard to explain but it really works so simply. To enjoy the process a little more I made this fun sun clock, check it out at https://sunsignal.app
I once experimented with beeswax candles as my only after-dark light source. This meant no hyper-stimulating screen activities whatsoever, too. TV, phone, video games, browsing the web? Nope, nope, nope, and nope. Just dim, warm light from actual flames.
Cured my lifelong “night owl” “trait” in a couple days. Shockingly effective.
Turned out to be hard to keep up and still, like, exist with other people, and you’d probably need to relax it a little in Winter unless your job lets you work reduced hours to kinda “hibernate” (otherwise when would you do anything that’s not work but requires light or electronics?) but it sure worked.
split jobs and managing people is hard.
Do does managnig agents.
Welcome to the club. Dealing with LLM is exhausting
LLMs poison your mind. The more AI slop you read, the more your mind turns into something like slop. This isn't very different from the idea that the food you eat is what your body is made of.
It sounds far fetched at first, but I think it could be true. There was a time in my life when I read literature constantly and started using those patterns in my writing. The habit fell off quickly after I stopped reading for a while, so there might be hope.
I feel this is coming for me. I now have three people generating code using LLMs who have no experience or real understanding of what they're doing. Nor do they have a strong desire to learn, but how could they anyway? I learnt by screwing up time and time again, but each time I went back and learnt why and next time did it better. It would be bad enough if I was managing 5 different agents myself, but it's managing 5 people who are operating 5 agents.
Speak for yourself! This is my third week on SSRI medication and I am the zombie the CEO wants me to be! /s (well, only half /s)
At work, I ended up doing other chores, getting a lot more involved in projects I wouldn't even care to touch. Turns out it's kinda fun being the source of truth at work. I now have a clear sketch of what the company has done and what we can improve on.
Being able to fill in the gaps at a company that doesn't do much feels like a company within a company. Sure, its not Silicon valley, but it's still fun! And job security is guaranteed if you do a bit more than just play the ticket factory