Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Generative AI coding tools and agents do not work for me

blog.miguelgrinberg.com

Post reply on HN
Print view View original

441–450 of 464 posts

Re: Generative AI coding tools and agents do not work for me

#441

Post by nsonha »

nottorp wrote: post #289
nsonha wrote: post #276

Earlier quoted context omitted.

> when Google used to do the same thing for you way before "AI"? Which is never? Do you often just lie to win arguments? LLM gives you a synthesized answer, search engine only returns what already exists. By definition it can not give you anything that is not a super obvious match

> Which is never? In my experience it was "a lot". Because my stack traces were mostly hardware related problems on arm linux in that period. But I suppose your stack traces were much different and superior and no one can have stack traces that are different from yours. The world is composed of just you and your project. > Do you often just lie to win arguments? I do not enjoy being accused of lying by someone stuck…

> your stack traces were much different and superior and no one can have stack traces that are different from yours

Very few devs bother to post stack traces (or generally any programming question) online. They only do that when they're stuck so badly.

Most people work out their problem then move on. If no one posts about it your search never hits.

Re: Generative AI coding tools and agents do not work for me

#442

Post by globnomulous »

kenjackson wrote: post #324
globnomulous wrote: post #189

Earlier quoted context omitted.

> I can read code much faster than I can write it. I have known and worked with many, many engineers across a wide range of skill levels. Not a single one has ever said or implied this, and in not one case have I ever found it to be true, least of all in my own case. I don't think it's humanly possible to read and understand code faster than you can write and understand it to the same degree of depth. The brain just…

You definitely can. For example I know x86. I can read it and understand it quite well. But if you asked me to write even a basic program in it, it would take me a considerable amount of time. The same goes with shell scripting. But more importantly you don’t have to understand code to the same degree and depth. When I read code I understand what the code is doing and if it looks correct. I’m not going over other des…

Huh, I don't know x86 but I do plenty of shell-scripting and am surprised, and a little embarrassed, that it had never dawned on me: you're right; they are easier to read, at least with a view to understanding intent, than to write. In fact, are there shell-scripting languages of which this isn't true?

Re: Generative AI coding tools and agents do not work for me

#443

Post by elbear »

nottorp wrote: post #336
elbear wrote: post #333

Earlier quoted context omitted.

It can also do that if you ask it. It can give you exercises that you can solve. But you have to specifically ask, because by default it just gives you code.

Of course, I originally was picking on Stack Overflow's moderation. Which strongly discouraged trying to teach people.

Oh, I missed that. I also missed that about StackOverflow, although I wasn't active there. It sounds like the motto of SO was "Solve problems!" instead of "Teach people how to solve problems!".

Re: Generative AI coding tools and agents do not work for me

#444

Post by nikanj »

elbear wrote: post #443
nottorp wrote: post #336

Earlier quoted context omitted.

Of course, I originally was picking on Stack Overflow's moderation. Which strongly discouraged trying to teach people.

Oh, I missed that. I also missed that about StackOverflow, although I wasn't active there. It sounds like the motto of SO was "Solve problems!" instead of "Teach people how to solve problems!".

Alas, someone bungled the incentive system and a closed question counts as a solved problem

Re: Generative AI coding tools and agents do not work for me

#445

Post by nottorp »

elbear wrote: post #443
nottorp wrote: post #336

Earlier quoted context omitted.

Of course, I originally was picking on Stack Overflow's moderation. Which strongly discouraged trying to teach people.

Oh, I missed that. I also missed that about StackOverflow, although I wasn't active there. It sounds like the motto of SO was "Solve problems!" instead of "Teach people how to solve problems!".

It kinda went downhill at some point. But currently:

> And ChatGPT never closes your question without answer because it (falsely) thinks it's a duplicate of a different question from 13 years ago

ChatGPT acts exactly opposite to the SO mods.

> But it does give you a ready to copy paste answer instead of a 'teach the man how to fish' answer.

Here it acts exactly like what SO mods like.

The other comments are mostly people thinking this is about ChatGPT...

Re: Generative AI coding tools and agents do not work for me

#446

Post by nottorp »

SpaceNugget wrote: post #366
nottorp wrote: post #289

Earlier quoted context omitted.

> Which is never? In my experience it was "a lot". Because my stack traces were mostly hardware related problems on arm linux in that period. But I suppose your stack traces were much different and superior and no one can have stack traces that are different from yours. The world is composed of just you and your project. > Do you often just lie to win arguments? I do not enjoy being accused of lying by someone stuck…

According to a quick search on google, which is not very useful these days, the maximum query length is 32 words or 2000 characters and change depending on which answer you trust. Whatever it is specifically, the idea that you could just paste a 600 line stack trace unmodified into google, especially "way before AI" and get pointed to the relevant bit for your exact problem is obviously untrue.

"is" not "was" though.

Pasting stack traces and kernel oopses hasn't worked in quite a while, I think. It's very possible that the maximum query was longer in the past.

2000 characters is also more than a double spaced manuscript page as defined by the book industry (which seems to be about 1500). You can fit the top of a stack trace in there. And if you're dealing with talking to hardware, the top can be enough.

Re: Generative AI coding tools and agents do not work for me

#447

Post by BlackFly »

blueflow wrote: post #261
BlackFly wrote: post #257

Earlier quoted context omitted.

One of the many ways that search got worse over time was the promotion of blog spam over actual documentation. Generally, I would rather have good API documentation or a user guide that leads me through the problem so that next time I know how to help myself. Reading through good API documentation often also educates you about the overall design and associated functionality that you may need to use later. Reading the…

You parent searches for answers, you search for documentation. Thats why AI works for him and not for you.

Did I say AI didn't work for me? Nope, so it is unclear to me why you are projecting some kind of perceived deficit onto me.

The difference between me and the person I responded to is that I feel I understand the perspective of the OP and I was trying to help the person who it didn't make sense to to understand the perspective.

Re: Generative AI coding tools and agents do not work for me

#448

Post by SpaceNugget »

nottorp wrote: post #446
SpaceNugget wrote: post #366

Earlier quoted context omitted.

According to a quick search on google, which is not very useful these days, the maximum query length is 32 words or 2000 characters and change depending on which answer you trust. Whatever it is specifically, the idea that you could just paste a 600 line stack trace unmodified into google, especially "way before AI" and get pointed to the relevant bit for your exact problem is obviously untrue.

"is" not "was" though. Pasting stack traces and kernel oopses hasn't worked in quite a while, I think. It's very possible that the maximum query was longer in the past. 2000 characters is also more than a double spaced manuscript page as defined by the book industry (which seems to be about 1500). You can fit the top of a stack trace in there. And if you're dealing with talking to hardware, the top can be enough.

I'm really struggling to see why you might assume that google decreases the maximum query length over time when that's generally the exact opposite of how things develop?

And indeed, in the early days the maximum query length was 10 words. So no, you have never been able to paste an entire stack trace into google and magically get a concise summary.

If you are changing the original claim that you were responding to to "I can do my job without llms if I have google search" Sure of course anyone can. But you can't use that to dismiss that some people find it quite convenient to just dump the entire stack trace into a text chat and have a decent summary of what is important without having to read a single part of it.

Re: Generative AI coding tools and agents do not work for me

#449

Post by dlock17 »

xpe wrote: post #340

> Unfortunately these claims are just based on the perception of the subjects themselves, so there is no hard data to back them up. Did the author take their own medicine and measure their own productivity?

Somehow you think the burden of proof is on his response rather than the AI maxis initial claims, but regardless, here you go. It was measured.

https://arxiv.org/abs/2506.08872

Re: Generative AI coding tools and agents do not work for me

#450

Post by jdbernard »

wiseowise wrote: post #166
jdbernard wrote: post #30

Earlier quoted context omitted.

If the AI actually outperforms humans in the full context of the work, then no, we won't. It will be so much cheaper and faster that businesses won't have to argue at all. Those that adopt them will massively outcompetes those that don't. However, assuming we are still having this conversation, that alone is proof to me that the AI is not that capable. We're several years into "replace all devs in six months." We wil…

> If the AI actually outperforms humans in the full context of the work, then no, we won't. IDEs outperform any “dumb” editor in full context of work. You don’t see any less posts about “I use Vim, btw” (and I say this as Vim user).

You're getting down-voted, I think, because you're missing the point. I would argue that the "I use Vim, btw" articles are themselves proof that there are still people for whom Vim is actually more performant. Similarly, you don't see people being passed over in hiring because of editor choice because it does but less to consistent results.

Compare to a hand saw. You still see them in specialty work and hobby shops, but you don't see them on construction sites. You see circular saws. Same with hammers. You'll probably still see them in job sites, but with far less usage than nail guns. And in many contexts nail guns have completely replaced hammers. There are still people griping about power tools but the industry doesn't care. I know a fair number of people in the trades and I can't imagine any of them seriously suggesting that you don't need to know how to use power tools.

My argument is that, assuming AI fulfills the expectation of those who hype it (and that assumption has yet to be proven), we will see a similar effect in software. The results will speak for themselves and make the arguments irrelevant. That hasn't happened yet, leaving room for genuine debate.

Post reply on HN