It’s irrelevant and unrelated.
vividfrier claims they haven’t written a line of code (implying other employees are similar), and their big company is operating normally. Bun is a big project and the rewrite is entirely LLM-generated. If its development continues normally, it reinforces the claim’s plausibility and proves someone made a large change (rewrite) entirely using AI. If not, it provides strong doubt: either vividfrier’s company is doing something different that avoids Bun’s problems (maybe other employees are still writing code manually), or they’re misleading or lying.
People write code differently, AI models write code differently, AI systems write code differently, companies create systems that write AI-written code differently, etc.
The system that wrote Bun bears no relationship to the system that writes OP's code.
Making such absolute statements about AI-written code is as dumb as making absolute statements about human-written code on the basis that it's "human-written".
It's plausible that OP's company is succeeding with 100% AI-generated code, even if Bun fails, but it's also plausibly false. Anyone can claim anything on the internet, what separates BS from reality is evidence.
I didn't write that Bun's rewrite absolutely proves or disproves OP's claim, I wrote that it provides evidence; it does, much more than OP's word.
It's also plausible that OP's claim is true, but only because despite being in a "big tech" company, they've been working on small self-contained repos, throwaway scripts, etc. The implications of this would be much different than what their comment suggests, which is another reason evidence matters: it forces them to narrow their claim, because anyone can make an overzealous claim from a small example.
That’s very different than converting a massive codebase one-off to an entire different language, while depending on tests to keep it contained.
Scale and process is dramatically different than the Bun case.
No need to look further.
Imagine old school machinists saying to a CNC machinist “Ha! See, maybe you don’t jog the axes manually, but you still have to be involved in placing the stock material, and you have to do the CAD/CAM work - so did it really machine the part for you? No!”
AI is a tool like any other. It has its limitations. It has classes of problems that it is suited to handle, and others it isn’t. If it’s true that they haven’t written (as in “typed out by hand”) a single line of code, why can’t they say that without you making that statement into more than it is?
I haven’t written a single line of code in 6 months, and that’s simply fact. It is also true that I put in a lot of other work to make that feasible, but that work isn’t in the form of writing code.
“it’s mature and the next step of engineering”
Tautologically, it’s mature enough for what it is mature enough for, and it certainly is the next step in the same way that CNC was the next step for machining — if you’re not using it as a machinist, you’re going to produce less compared to those who are.
Same thing with garden hoses. Yes, you can go fetch water from a lake and splash it on your lawn, or, you know, you could just use a sprinkler connected to your garden hose. Doesn’t replace buckets. Buckets just have a narrower scope in a world where garden hoses exist.
It also had a logical stopping point in automation tech.
Ai is trying to do everything and wont stop
Because it's a solution looking for a problem. All the AI companies lean in to coding because it actually helps with that to some degree but the amount that it helps doesn't justify their valuations. It needs to be good at everything to justify their target IPO price.
A garden hose vs a bucket is also the same situation. You can accomplish the same thing with either, but one might be more labor intensive.
AI is nothing like either of those. It would be like instead of a bucket you get a garden hose that points in a different direction every time you try to use it. Or instead of a 5 axis mill that rigorously executes the g-code it just randomly reinterprets tool paths each time it cuts a part. Both of these things would be worse than useless in their respective applications.
AI is different because it plays to the pliability of the software domain. Even fairly shitty, irreproducible results can be good enough for software development, if you don't look at it too closely. Make analogies to the physical world at your peril!
And also adds a multiplier to your water bill
The same with AI you still have to hold it and point in direction to be useful.
A common one: "I have stopped writing code, the world is going to end"
Another: "I will code by hand, I don't care"
Another: "I use it as a tool, but the hype bothers me so much that I have to bitch and moan from morning to night"
This one is: "I have stopped writing code, it wasn't the end of the world."
You can use agents for dev and reduce MTBF
Examples of hype right now in SV (could happen, but more evidence needed imo):
- RSI
- Dark factories
- Overnight swarms, full rewrite one-shots
- 1 person unicorn
- AI native services co
- UBI / UB compute
- "permanent underclass"
Usually they provide grandiose claims (like the top-level comment) without any evidence or just anecdotal evidence that is not verifiable.
HN is lousy with new accounts (created in the past year) that are overwhelmingly excited for the so-called AI revolution.
Oh look more useless arguing.
People who do things care about the doing more than how the sausage was made.
I do not care how software gets built. Only that it works. Results is the only thing that matters and I hope everyone in this thread internalizes that fact.
I mean, I agree on a very high level of abstraction. But my problem is that I need to understand how software gets built so that I can have confidence in my ability to maintain and evolve the project.
I need to understand whether a feature is easy to add or requires a wholesale rewrite of the entire codebase, which comes with risks. I need to understand how new features affect existing users.
I also need to understand the economics of the process and the economics of my industry. That means I have to care to some degree about how software gets made, not just whether some specific program works at the present moment.
If you give me a choice between an implementation that is 100 LOC I can understand and an implementation that is a million LOC that I can never understand, I'm going to chose the former, even if both implementations pass all tests.
this is not a thing. there hasn't been a single line of code written that a human can't understand.
Also, code quality matters for AI as well. Maintaining a million lines of code requires more tokens than maintaining 100 lines of code.
Then do it. Be successful. Be wonderful. Show us all the great results.
If you only care about results, then go do the things.
Why are you complaining? Whats wrong with differing views?
When people cannot accept other people critiquing AI; that is literally AI psychosis.
Heres a trick: what is AI bad at? Stop and ask yourself what it really sucks at.
Nothing come to mind?
You're living AI psychosis.
If you can't accept that anything it does is wrong or bad, that you are only successful when you use AI you are, flat out, gas lighting yourself.
Now go read the comments by recent new accounts about AI.
Yeah.
Either there are a lot of bots, or a lot of really really troubled people out there right now.
This account is my real name. I have nothing to hide. The metrics speak for themselves.
We just had a week with 7 major zero days announced for pretty much every major OS and architecture. Reality doesn't care about your opinion.
What metrics? Where are the amazing new projects and features you built? Where are the amazing products and features you built that are better than existing ones (run faster, consume fewer resources etc.)?
For a person who "has nothing to hide" somehow none of your comments ever mention what projects you work on, what you ship, or what metrics you employ.
New php extension https://github.com/hparadiz/ext-gnu-grep
A Demo showing how to stream webrtc to KDE Wayland overlay. https://github.com/hparadiz/camera-notif
A fun little tool that captures stdout/stderr on any running process. https://github.com/hparadiz/bpf_write_monitor
Then I upgraded my 10 year old hand written framework to a new version that supports sqlite and postgres on top of existing MySQL support https://github.com/Divergence/framework
But then I was like eh lemme benchmark every PHP orm that exists just to check my framework's orm....
https://github.com/hparadiz/the-php-bench
And published the results.... Here
https://the-php-bench.technex.us/
And then I decided to vibe code a simulation of the entire local steller group https://earth.technex.us
Followed by my simulation of the Artemis 3 landing sites at the lunar South pole https://artemis-iii.technex.us/?scale=1.000#South-Pole
And I left the best for last.....
https://github.com/hparadiz/evemon
A brand new task manager written in C for Linux that supports a plugin architecture with an event bus. It's literally the best gui Linux task manager ever. Still working on it.
I'm not even talking about my paid job. This is me just fucking around.
What I said was: what is AI bad at?
Let’s start the conversation there. Can you one shot a new Linux kernel with AI? Does “new project, are me GTA 8!” Work?
No.
That is hard reality, disconnected from any ridiculous hype BS people are living in.
You need to start the conversation with: I acknowledge, AI is not good at some things.
Now, if you can accept that baseline reality that you live in you can ask some hard questions like:
Are the things I’m doing things I could do without AI?
Am I actually more productive?
Is what I’m building actually working?
Are the things that other people are claiming to do, things that it can actually do?
Is it the right tool for the next thing I’m going to work on?
Maybe you don’t care; but, if you don’t like people asking those questions; why?
This is genuinely great tech. No one is denying that; it’s effective and productive to get certain stuff done.
What scares me is people who you cant even have this conversation with. What’s AI bad that?
Nothing!
Thats AI psychosis.
Take a look at your own opinions and why you take it personally (it seems) when someone calls out a case where AI did a bad job, or someone who failed because they relied too hard on AI.
Seriously man, if those things are upsetting you, because, literally, they have nothing to do with you, other than the tiny voice in your head telling you that “critiquing AI is critiquing me”… then have a good think about that.
It’s not a good place to be.
Build things. Be happy, be proud. Coming here and ranting like you’re doing? It’s not good. It’s realllly not good.
--- start quote ---
- Just what kind of evidence do you suppose they could have?
- Showing actual improved products and features. Showing actual code. etc.
--- end quote ---
Note how you provided neither. It's just claims.
> Anyway AI for sure increased project cadence by at least 2x.
As in: you claim this. Also, no one denies that you can ship a lot of code much faster with AI. However, somehow, very little actual evidence of grandiose claims (see farther up in the context) besides anecdotal "I'm so faster and features are being shipped left and right".
See also a sibling comment: https://news.ycombinator.com/item?id=48158565
I think it's a great example of using LLMs effectively. In the end becoming more productive is understanding where LLMs work great and where they fail miserably.
But it is a step similar to, say going from assembly to a higher-level programming language, not the silver bullet that AI astroturfers like you to believe (fire all the programmers to buy more tokens!)
I’m currently working on a rewrite of an app that originally took two years. It’s been about three months, and I’m probably about 70% done. It’s a total “from scratch” rewrite; both client and server (two versions of each, as I also have administrative code). It’s a pretty big system, for one guy. I couldn’t do it, without the LLM.
It’s not been a cakewalk. I’ve needed to toss out large swaths of LLM-generated code, and rewrite by hand, but, for the most part, it’s been a huge help.
But I’m also not doing it in a manner that eats tokens. I just use the standard $20/month subscription as a chat. I suspect my workflow is not one that Anthropic or OpenAI really wants out there.
But I also bet that many HN accounts are bots; although I think many may be ones run by enthusiasts, not some AI cabal.
Other than that, I am not boosting AI, and have absolutely zero interest in doing a bunch of work to satisfy some random Internet Guy, who can't be bothered to examine my pretty damn extensive open portfolio.
I was just talking about my personal experience.
> random Internet Guy, who can't be bothered to examine my pretty damn extensive open portfolio.
You cannot even be bothered to examine the comment you reply to, maybe get off your high horse.
And the main part of my comment was about something in the common realm, open source software, and hard performance/quality improvements. Not wishy-washy products and features, not yet another tone deaf cool story.
For instance, I checked out yours, and there's not much, except a whole bunch of challenging people here. I am wondering if you came here to "set us straight." I know that a lot of folks have low opinions of HN, and not all of them are wrong, but I find this place a fairly good place to hang out. Being challenged, is one of the draws, for me.
By the way, have you tried the new unhomogenized heavy cream? Good stuff!
Have a great day!
Not everyone is you. E.g. I don't expect he answer to "show your work" be no answer and "why didn't you check my profile".
My answer to "show your work" was "No." I am not going to go through my code, and show a bunch of supporting evidence for a casual comment, in which I have exactly zero investment. I really don't care that much what people think of me. I was just sharing my personal experience. If you guys want to write me off, then knock yourselves out.
"No" is a complete sentence. What part of "No" didn't he understand?
Have a great day!
An interesting answer to literally "Just what kind of evidence do you suppose they could have? - Showing actual improved products and features. Showing actual code. etc."
> "No" is a complete sentence. What part of "No" didn't he understand?
See above. After pointing this out you immediately started down the path of "why didn't you looko at my profile and followed the link to my github".
Not even close. The opposite even. Read more carefully.
> For 5 million comments like yours I haven't seen a single one with the old code vs. the new code. I understand that not all code is public that way, of course, and I don't mean to put you on the spot personally. But where are all the open source projects that now do the same with better error handling using less resources?
Same here :)
> not some AI cabal.
There are enough enthusiasts to make it feel like one. Also an unhealthy doze of marketers, people buying into hype, AI psychosis etc.
There's absolutely no question that AI is a real thing, and that there's going to be a lot of money made, so there's a bunch of folks with commercial interest in pushing it.
It's just different from crypto. This has actual real-world utility for just about everyone. I am increasingly hearing people say "Ask ChatGPT," where they used to say "Google It" (where they used to say "Look it Up at the Library").
But your anecdote is much more balanced and more in line with my personal findings. Not like all the AI astroturfing that happens here. I like to use LLMs as well, but in a very targeted way in places where LLMs shine.
Yes, LLMs can make you much more productive. But so could assembly -> C -> Python or Rust, or switching an IDE with code completion and support for refactoring. Each step makes you more productive.
Sure, LLMs can spit out greenfields projects. But on large projects with complex requirements, you still need senior engineers to guide them, carefully review the output, and as you say throw out code and write it from scratch in a better way.
I had some friends who ended up in bits of AI psychosis. They exclaim that a swarm of agents was writing all their code, but every time I ask them to show the end-result, all they have is a pile of code that they don't understand, nor doesn't really work either. At the same time, they stopped getting any actual work done.
At any rate, somebody had a great analogy on HN recently: think of it is a vector, LLMs can significantly increase the magnitude of the vector, but you still have to make sure that the orientation of the vector is correct.
Great analogy!
https://www.youtube.com/watch?v=AkKo1_RP_0c
The comment you’re replying to wasn’t uncivil. It wasn’t rude. It was a lament.
I’m not advocating for this rule to change (I’d appreciate if you didn’t straw man and mischaracterise what I said), but I am saying if a problem happens over and over and people notice it and talk about it, then you should maybe pay attention. The rule for new accounts came about from multiple comments and even submissions asking for it, not private emails. It came about from community conversation and outcry.
The load-bearing word in that claim is "undue", and it's not justified here. I'm not doing arcane rules-lawyering, I'm just saying people should avoid doing things the site guidelines quite specifically ask them not to do.
> I’m not advocating for this rule to change (I’d appreciate if you didn’t straw man and mischaracterise what I said),
I wasn't suggesting you did, I was suggesting the person I originally replied to might.
https://en.wikipedia.org/wiki/Generic_you
Does that mean I now repeat your parenthetical back to you? ;)
Which I agree with. And I’m just saying the rules aren’t absolute, can’t cover every situation, and could not predict the change of the world around them, thus occasional deviation from them is OK, especially when it serves the larger goal of protecting discourse on a website whose rules were written to protect it.
It’s the spirit of the law VS the letter of the law. Let’s say the rules ask you to not shove people but say nothing about peeing on others. If someone suddenly starts peeing on everyone without consent and refuses to stop, shoving them to get them away becomes an appropriate response despite being technically against the rules.
> I wasn't suggesting you did, I was suggesting the person I originally replied to might. (…) Does that mean I now repeat your parenthetical back to you?
By your own logic, I wasn’t suggesting you did it, I was asking for no one else to do it (also, it’d make no sense anyway, it’s not a straw man to incorrectly say someone is straw manning). It also means the person you originally replied to wasn’t accusing their parent comment, thus making your original comment invalid.
Also I provided the list of actual hype ha
Often such comments appear just before the submission is abandoned to wrap up the thing.
> We still require a code review for any change and it's becoming a bottleneck - for sure
> I fought for years trying to convince my colleagues to write good commit messages. Now Claude is writing great commit messages but since *I'm no longer looking at code* - I never see them. I don't think Claude uses them either.
(emphasis added)
In a way, obvious injected opinions benefit culture, by making formerly-unaware readers skeptical.
I have almost certainty by now that half of the web is just bots steering the narrative of humans, because I’ve never seen so much non sense being normalized in my life. Dish soap drinking level I mean.
Seriously, not a bot. Just a software engineer who feels gaslighted because I see AI used in one way at work and then I go to Hacker News and I feel like I am using a different technology compared to everyone else.
And don't get me wrong. I would like to be more skeptic about AI because I enjoy writing code, I enjoy a high salary with great benefits. But with the speed and direction we're seeing now (so I am not only looking at this specific point in time but rather also the direction and the fact that no one knows what they're doing) - I do worry about losing this. So I am definitely crossing my fingers that it is a bubble but I just don't see the evidence yet.
Above, you said:
> We still require a code review for any change
And:
> We don't really vibe though. At least I don't. I see it more as comment driven development. I need to understand the code and what I want to achieve where in the codebase
But in your previous comment, you said:
> since I'm no longer looking at code
And:
> Branches are now irrelevant
How can all of these things be true?
Lots of tech companies are doing just fine with purely AI written code at this point.
Saying that the quality is getting worse in some immeasurable way (while incidents remain the same) is literally unfalsifiable.
If there are so many, surely you or one of the other AI supporters have a public example?
I’m aware of two examples, although they’re (mostly routine) translation with existing test infrastructure, so easier for an LLM:
- Bun’s rewrite, although we haven’t seen the effects on further development
- Ladybird’s rewrite, which seems to be continuing fine
I don't know what the Bay Area note is supposed to mean in the context of the whole post - unless you want to reinforce that it surely means that it's a sane take... In which case, I'm not certain the non-Bay readers would agree that it comes from an unbiased culture.
1) Since vibe coding GitHub has frequent outages and isn't able to load a large number of comments.
2) The slop translation of Bun resulted in immediate bugs (https://github.com/oven-sh/bun/issues/30719) that the hyped Mythos apparently did not find.
3) AI features and (likely, though not proven) AI code resulted in a 0-day in Google code:
https://projectzero.google/2026/01/pixel-0-click-part-1.html
The house of cards is beginning to collapse.
What we do see publicly is OSS projects overrun with poor submissions, for example.
It is like saying "no one can produce a viable CPU because I can't tape one out in my garage."
Focus on that message, not the messenger.
HN was a tremendous resource built by its members and the moderators. In the last year or so a lot of that has been destroyed by people who have no sense of decency. They see deception as a virtue. They call it hustle or whatever. WTF?
Pretty sus, bot or otherwise.
Just like I need to keep in mind that not everyone work in a big tech(ish) company where tech stacks are standardised and many problems are solved centrally, I feel like people also need to keep in mind that not everyone has a horrible experience with AI.
And I also want to clarify that I hate AI. I loved writing code, I loved having a comfortable job with a good salary and good benefits. While the company I work for haven't done "AI layoffs" yet I feel like it's a matter of time because I'm not only looking at the state we are right now but also what the direction and speed is. It was only back in October I felt like AI was a bubble about to burst cause I wasn't getting enough value out of Claude Code. And then ofc Opus 4.5 happened and it changed my view completely. And we're a few months into using 4.5/4.6/4.7 (and I'm not an Anthropic shill, currently I am using Codex most of the time) but I was hoping this career would last me decades. But where will be 5 years from now? 15 years from now?
I just want to provide my perspective from a big tech company where I feel AI is doing more and more parts of my job. It's not perfect and it gets things wrong but I mean... humans do as well?
And below you repeat what all of Hacker News hypemen say about AI (“I have stopped writing code”, “it’s mature and the next step of engineering”)
Thank you for reinforcing the point of OP
EDIT: you're the same person that a month ago said your company feels git is outdated now that you have agentic coding, and you don't even need to write your own commit messages. This is next-level trolling, or a serious case of AI psychosis.