Live data from Hacker News

The hidden cost of AI coding

terriblesoftware.org

311–320 of 475 posts

Re: The hidden cost of AI coding

#311
post #246

Earlier quoted context omitted.

> On the other hand, I know people that want to jump straight to the end result. They have some melody or idea in their head, and they just want to generate some song that revolves around that idea. I don't really look down on those people, even though the snobs might argue that they're not "real musicians". I don't understand them, but that's not really something I have to understand either. So if someone generates…

Where do you draw that line though? If you're talking about a person using an LLM, or some other ML system, to help generate their music then the LLM is really just a tool for that person. I can't run 80 mph but I can drive a car that fast, its my tool to get the job done. Should I not be allowed to do that professionally if I'm not actually the one achieving that speed or carrying capacity? Personally my concerns wi…

> I can't run 80 mph but I can drive a car that fast, its my tool to get the job done.

Right, but if you use a chess engine to win a chess championship or if you use a motor to win a cycling championship, you would be disqualified because getting the job done is not the point of the exercise.

Art is (or should be) about establishing dialogues and connections between humans. To me, auto-generated art it's like choosing between seeing a phone picture of someone's baby and a stock photo picture of a random one - the second one might "get the job done" much better, but if there's no personal connection then what's the point?

Re: The hidden cost of AI coding

#312

I have actually had some really great flow evenings lately, the likes of which I have not enjoyed in many years, precisely because of AI-assisted coding. The trick is to break the task down in to components that are of moderate complexity so that the AI can handle them (Gemini 2.5 Pro one-shots), and keep your mind on the high-level design which today's AI cannot coordinate. What helps me is to think of it like I'm a…

people seem to forget this type of argument from the article was used for stack overflow for years, calling it the destruction of programming. "How can you get into flow when you are just copying and pasting?". Those same people are now all sour grapes for AI assisted development. There will always be detractors saying that the documentation you are using is wrong, the tools that you are using are wrong, and the meth…

The issue with "AIOP" is that you don't have a litany of others (as is the case with SO) providing counter examples, opinions, updated best practices, etc. People take the AI output as gospel and suffer for it without being exposed so the ambiguity that surrounds implementing things.

Re: The hidden cost of AI coding

#313

Earlier quoted context omitted.

Good artists copy, great artists steal

Picasso (if he really said that) had a machine painting for him?

Picasso explicitly wanted his designs (for cutlery, plates, household items he designed) to be mass-produced, so your question is not as straightforward as you make it to be.

Re: The hidden cost of AI coding

#314

Earlier quoted context omitted.

i mean how far are you willing to take that argument? every decade has just been a new abstraction, imagine people flipping switches or in raw assembly talking about how they don't "understand" you now with your no effort. or even those who don't "understand" why you use your autocomplete and fancy IDE, preferring a simple text editor. i say this as someone who cut my teeth on this stuff growing up and seeing the evo…

It is not just gatekeeping. It is a stubborn refusal to see that one could be programming something much more sophisticated if they could use these iteration loops efficiently. Many of these folks would do well to walk over to the intersection of Market, Bush, and Battery Streets in San Francisco and gaze up at the Mechanics Monument.

> It is a stubborn refusal to see that one could be programming something much more sophisticated if they could use these iteration loops efficiently

Programming something more sophisticated with AI? AI is pretty much useless if you're doing anything somewhat novel. What it excels at is vomiting code that has already been written a million times so you can build yet another Electron cross-platform app.

Re: The hidden cost of AI coding

#315

I have actually had some really great flow evenings lately, the likes of which I have not enjoyed in many years, precisely because of AI-assisted coding. The trick is to break the task down in to components that are of moderate complexity so that the AI can handle them (Gemini 2.5 Pro one-shots), and keep your mind on the high-level design which today's AI cannot coordinate. What helps me is to think of it like I'm a…

I always wonder what kind of projects are we talking about. I am currently writing a compiler and simulation engine for differential-algebraic equations. I tried few models, hoping they would help me, but they could not provide any help with small details nor with bigger building blocks. I guess if you code stuff that had been coded a lot in public repos, it is fine, otherwise AI does not help in any way. Actually, I…

That's been my experience. If it's been solved a million times, it's helpful. If you're out on the frontier where there's no public code, it's worse than useless.

If you're somewhere in between (where I am now) it's situationally useful for small sub-components but you need to filter it heavily or you'll end up wasting a day or two going down a wrong rabbit-hole either because you don't know the domain well enough to tell when it's bullshitting or going down a wrong path, or don't know the domain well enough to use the right keyword to get it to cough up something useful. I've found domain knowledge essential for deciding when it's doing something obviously wrong instead of saying "I don't know" or "This is the wrong approach to the problem".

For the correct self-contained class or block of code, it is much faster to specify the requirements and go through a round or two of refinement than it is to write it myself. For the wrong block of code it's a complete waste of time. I've experienced both in the last few days.

Re: The hidden cost of AI coding

#316

Most comments here surprise me: I am using Githubs Copilot / ChatGPT 4.0 at work with a code base which is mostly implements a basic CRUD service... and outside of small/trivial example (where the generated code is mostly okay), prompting is more often than not a total waste of time. Now, I wonder if I am just totally unable to write/refine good prompts for the LLM (as it works for smaller samples, I hope I am not to…

> Now, I wonder if I am just totally unable to write/refine good prompts for the LLM (as it works for smaller samples, I hope I am not too far off) or what could explain the huge discrepancy of experience.

Programming language / stack plays plays a big role, I presume.

Re: The hidden cost of AI coding

#317

Some people love programming, for the sake of programming itself. They love the CS theory, they love the tooling, they love most everything about it. Other people see all that as an means to an end - and find no joy from the technical aspect of creating something. They're more interested in the end result / product, rather than the process itself. I think that if you're in group A, it can be difficult to understand g…

I’m in group A and B. I do programming for the sake for it at home. I read tons of technical books for the love of it. At work, though, I do whatever the company wants or whatever they allow me… I just do it for the money.

Re: The hidden cost of AI coding

#318
post #8

Earlier quoted context omitted.

So much this. The AI takes care of the tedious line by line what’s-the-name-of-that-stdlib-function parts (and most of the tedious test-writing parts) and lets me focus on the interesting bits like what it is I want to build and how the pieces should fit together. And debugging, which I find satisfying. Sadly, I find it sorely lacking at dealing with build systems and that particular type of boilerplate, mostly becau…

> The AI takes care of the tedious line by line what’s-the-name-of-that-stdlib-function parts (and most of the tedious test-writing parts) AI generated tests are a bad idea.

Just as with anything else AI, you never accept test code without reviewing it. And often it needs debugging. But it handles about 90% of it correctly and saves a lot of time and aggravation.

Re: The hidden cost of AI coding

#319
post #138
post #8

Earlier quoted context omitted.

So much this. The AI takes care of the tedious line by line what’s-the-name-of-that-stdlib-function parts (and most of the tedious test-writing parts) and lets me focus on the interesting bits like what it is I want to build and how the pieces should fit together. And debugging, which I find satisfying. Sadly, I find it sorely lacking at dealing with build systems and that particular type of boilerplate, mostly becau…

Aren't stdlib functions the ones you know by heart after a while anyways?

Depends on the language. Python for instance has a massive default library, and there are entire modules I use anywhere from one a year to once a decade —- or never at all until some new project needs them.

Re: The hidden cost of AI coding

#320

Earlier quoted context omitted.

They perhaps haven’t taken away your keyboard but anecdotally, a few friends work at places where their boss is requiring them to use the LLMs. So you may not have to code with them but some people are starting to be chained to them.

Yes, there are bad places to work. There are also places that require detailed time tracking, do not allow any time to write tests, have very long hours, tons of on-call alerts, etc.

How long until it becomes the rule because of some arbitrary "productivity" metric? Sure, you may not be forced to use it, but you'll be fire for being "unproductive".
Post reply on HN