I'm embracing the AI future we live in. It should be possible to build great software with AI. AI is good at code generation, but not testing. That's the other part of the equation we seem to be forgetting.
With the first and only commit 2 hours ago, the author of this project didn't let it "bake". They didn't exercise it locally to see what issues it might have, and I have a hard time believing the very first iteration of this software is perfect. With how easy it is to prompt/push anything, I'm not interested in engaging with anything that hasn't aged a bit.
> With the first and only commit 2 hours ago, the author of this project didn't let it "bake". They didn't exercise it locally to see what issues it might have, and I have a hard time believing the very first iteration of this software is perfect.
You have no idea whatsoever how many iterations were done before the initial commit. The VCS log is not representative of anything that happened before the first public release. Even before LLMs, people would grind away on stuff until they were happy and then put it in a VCS for public consumption.
Off topic, but watching the output of this thing renews my disappointment that so many programming languages/frameworks fail to provide signal sender information to signal handlers. It’s there! Crusty ABI and historical reasons are, as far as I can tell, the only reason why languages like Python (and many others) don’t surface sigaction to handlers without unnecessary acrobatics.
Sigh. We shouldn’t have modeled signal handling in a cross-platform, lowest-common-denominator way in HLLs. Signals have way less in common across OSes than files/sockets and the like.
Anyway, rant over. Not much to be done about it now.
Thank you, but I won't even consider a yet another AI/LLM slop which will not be maintained in a few months, nor anyone will ever invest their life time into, too.
> Hey everyone, I wrote this tool...
Hey there. And yes, I don't you believe it's you who "wrote" it (where even the Readme file seems to be generated), nor you have experience enough in the so crucial subjects raised, to invest my own life time in the project, too, sorry.
I'll better consider projects where actual effort and human was involved believing in their art, knowledge, and experience of life they express in their actual, authentic, and accountable work.
We put a lot of work into abstracting BPF into a JavaScript framework to give builders, human or agentic, the ability to use complex kernel primitives with a familiar programming model to build production grade, scalable infrastructure tools.
Our hope is that everyone here will soon be able to build their own tools on-demand, versus buying them.
We separate a lot of the data layer code from the presentation-level code, to make them re-usable and keep the test surface on the actual data's accuracy manageable across all the random edge cases that arise across CPU architectures, kernel versions, and environments like AWS ECS network namespace voo-doo.
I can relate to the tilting you're on, but try to consider the good that can come from this age, where such a tool might not have been built otherwise. It's also BSD/GPL license.
As an aside, I've been vibe-cooking for a few months on a personal project that's accomplished something lovely. For me. I sometimes wonder if I should give it away much like this project. But public reactions like yours temper the thought.
> Thank you, but I won't even consider a yet another AI/LLM slop which will not be maintained in a few months, nor anyone will ever invest their life time into, too.
This applied to most of my side projects before AI. Most of them I would never touch again.
Thanks to AI I'm working on them way more, and at a much higher level of engineering standards (especially the recent models are voluntarily adding tests, looking for bugs etc.).
(Well, except for the part about barely reading the code, but I said higher, not high!)
Also I realized the other day that I already reached the point where I don't understand my own code, several years before involving AI in the process!
I don't know if I'm an outlier but I thought that was pretty funny.
I'm embracing the AI future we live in. It should be possible to build great software with AI. AI is good at code generation, but not testing. That's the other part of the equation we seem to be forgetting.
With the first and only commit 2 hours ago, the author of this project didn't let it "bake". They didn't exercise it locally to see what issues it might have, and I have a hard time believing the very first iteration of this software is perfect. With how easy it is to prompt/push anything, I'm not interested in engaging with anything that hasn't aged a bit.
> With the first and only commit 2 hours ago, the author of this project didn't let it "bake". They didn't exercise it locally to see what issues it might have, and I have a hard time believing the very first iteration of this software is perfect.
You have no idea whatsoever how many iterations were done before the initial commit. The VCS log is not representative of anything that happened before the first public release. Even before LLMs, people would grind away on stuff until they were happy and then put it in a VCS for public consumption.
True. But it is still more likely than not that this project has seen limited testing on only a handful of machines.
I won't begrudge anyone who feels like it's too big a risk to engage with just yet.
I think being able to do this in a couple hours with AI is the point. He clearly used the yeet runtime to make it.
Off topic, but watching the output of this thing renews my disappointment that so many programming languages/frameworks fail to provide signal sender information to signal handlers. It’s there! Crusty ABI and historical reasons are, as far as I can tell, the only reason why languages like Python (and many others) don’t surface sigaction to handlers without unnecessary acrobatics.
Sigh. We shouldn’t have modeled signal handling in a cross-platform, lowest-common-denominator way in HLLs. Signals have way less in common across OSes than files/sockets and the like.
Anyway, rant over. Not much to be done about it now.
Thank you, but I won't even consider a yet another AI/LLM slop which will not be maintained in a few months, nor anyone will ever invest their life time into, too.
> Hey everyone, I wrote this tool...
Hey there. And yes, I don't you believe it's you who "wrote" it (where even the Readme file seems to be generated), nor you have experience enough in the so crucial subjects raised, to invest my own life time in the project, too, sorry.
I'll better consider projects where actual effort and human was involved believing in their art, knowledge, and experience of life they express in their actual, authentic, and accountable work.
Hey, I'm the founder of yeet.
This tool was built on top of our engine: yeetd.
We put a lot of work into abstracting BPF into a JavaScript framework to give builders, human or agentic, the ability to use complex kernel primitives with a familiar programming model to build production grade, scalable infrastructure tools.
Our hope is that everyone here will soon be able to build their own tools on-demand, versus buying them.
We separate a lot of the data layer code from the presentation-level code, to make them re-usable and keep the test surface on the actual data's accuracy manageable across all the random edge cases that arise across CPU architectures, kernel versions, and environments like AWS ECS network namespace voo-doo.
I can relate to the tilting you're on, but try to consider the good that can come from this age, where such a tool might not have been built otherwise. It's also BSD/GPL license.
As an aside, I've been vibe-cooking for a few months on a personal project that's accomplished something lovely. For me. I sometimes wonder if I should give it away much like this project. But public reactions like yours temper the thought.
It seems like a tool built for this bigger project[1] which actually looks kind of creative and cool.
[1] https://yeet.cx/
> Thank you, but I won't even consider a yet another AI/LLM slop which will not be maintained in a few months, nor anyone will ever invest their life time into, too.
This applied to most of my side projects before AI. Most of them I would never touch again.
Thanks to AI I'm working on them way more, and at a much higher level of engineering standards (especially the recent models are voluntarily adding tests, looking for bugs etc.).
(Well, except for the part about barely reading the code, but I said higher, not high!)
Also I realized the other day that I already reached the point where I don't understand my own code, several years before involving AI in the process!
I don't know if I'm an outlier but I thought that was pretty funny.