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

Reports of code's death are greatly exaggerated

stevekrouse.com

Post reply on HN
Print view View original

291–300 of 486 posts

Re: Reports of code's death are greatly exaggerated

#291

Post by jimbob45 »

elgertam wrote: post #174
lateforwork wrote: post #71

Chris Lattner, inventor of the Swift programming language recently took a look at a compiler entirely written by Claude AI. Lattner found nothing innovative in the code generated by AI [1]. And this is why humans will be needed to advance the state of the art. AI tends to accept conventional wisdom. Because of this, it struggles with genuine critical thinking and cannot independently advance the state of the art. AI…

You know where LLMs boost me the most? When I need to integrate a bunch of systems together, each with their own sets of documentation. Instead of spending hours getting two or three systems to integrate with mine with the proper OAuth scopes or SAML and so on, an LLM can get me working integrations in a short time. None of that is ever going to be innovative; it's purely an exercise in perseverance as an engineer to…

Couldn’t agree more and especially when some docs are incorrect and AI is able to guesstimate the correction based on other implementations or parallel docs that it’s found. Goes from “Let me spend a few days scouring the internet and our internal repo to see if I can maybe find a workaround” to “This can definitely get done”.

Re: Reports of code's death are greatly exaggerated

#292

Post by ssl-3 »

blackqueeriroh wrote: post #265
komali2 wrote: post #224

Earlier quoted context omitted.

> Business problems are essentially neverending That feels overly optimistic. LLMs seems on track to automate out basically any "email job" or "spreadsheet job," in which case we'll be looking at higher unemployment numbers than the great depression for at least some period of time. Combine with increased automation... There are a LOT of people in the world and already a not insignificant portion can't find work desp…

Do you really think the billionaires are willing to have consumers so impoverished that they can’t continue to spend large sums of discretionary income buying the things that make the billionaires themselves richer?

The impoverished are cheaper to enslave.

Re: Reports of code's death are greatly exaggerated

#293

Post by lionkor »

echelon wrote: post #140
lionkor wrote: post #136

To all the vibe coders: When you let an LLM author code, it takes ownership of that code (in the engineering sense). When you're done spending millions on tokens, years of development, prompt fine tuning, model fine tuning, and made the AI vendor the fattest wad of cash ever seen, you know what the vendor will do? You have to migration path. Your Codex prompts don't work the same in Claude. All the prompts you develo…

> When you're done spending millions on tokens, years of development, prompt fine tuning, model fine tuning, and made the AI vendor the fattest wad of cash ever seen, you know what the vendor will do? They'll hire the person who knows AI, not the human clinging onto claims of artisanal character by character code. It's entirely possible to engineer well-designed and intentional systems with AI tools and not stochasti…

It's almost like I addressed my entire comment to vibe coders and NOBODY else, because other uses of AI are pretty valid

Re: Reports of code's death are greatly exaggerated

#294

Post by lionkor »

thi2 wrote: post #156
lionkor wrote: post #136

To all the vibe coders: When you let an LLM author code, it takes ownership of that code (in the engineering sense). When you're done spending millions on tokens, years of development, prompt fine tuning, model fine tuning, and made the AI vendor the fattest wad of cash ever seen, you know what the vendor will do? You have to migration path. Your Codex prompts don't work the same in Claude. All the prompts you develo…

It is the same as adding dependencies or hosting on Azure/Aws, choosing a nosql db isn't it?

Well, except it's your entire codebase, yeah

Re: Reports of code's death are greatly exaggerated

#295

Post by vidarh »

lateforwork wrote: post #71

Chris Lattner, inventor of the Swift programming language recently took a look at a compiler entirely written by Claude AI. Lattner found nothing innovative in the code generated by AI [1]. And this is why humans will be needed to advance the state of the art. AI tends to accept conventional wisdom. Because of this, it struggles with genuine critical thinking and cannot independently advance the state of the art. AI…

You won't find anything innovative in most human-written compilers either, so by that argument we can't advance the state of the set either.

Re: Reports of code's death are greatly exaggerated

#296

Post by steve_adams_86 »

vidarh wrote: post #295
lateforwork wrote: post #71

Chris Lattner, inventor of the Swift programming language recently took a look at a compiler entirely written by Claude AI. Lattner found nothing innovative in the code generated by AI [1]. And this is why humans will be needed to advance the state of the art. AI tends to accept conventional wisdom. Because of this, it struggles with genuine critical thinking and cannot independently advance the state of the art. AI…

You won't find anything innovative in most human-written compilers either, so by that argument we can't advance the state of the set either.

We created compilers in the first place. I suppose an interesting question is: would LLMs have come up with compilers if humans hadn't?

Re: Reports of code's death are greatly exaggerated

#297

Post by vidarh »

steve_adams_86 wrote: post #296
vidarh wrote: post #295

Earlier quoted context omitted.

You won't find anything innovative in most human-written compilers either, so by that argument we can't advance the state of the set either.

We created compilers in the first place. I suppose an interesting question is: would LLMs have come up with compilers if humans hadn't?

"We", yes, but my point is that most people who write compilers do nothing but implement known techniques. If you then judge human ability to innovate by investigating a single compiler for innovation, odds are you would get the entirely wrong idea of what we are capable of.

Re: Reports of code's death are greatly exaggerated

#298

Post by scrollaway »

hrmtst93837 wrote: post #289
stalfie wrote: post #86

Earlier quoted context omitted.

Well, to be fair, judging by the shift in the general vibes of the average HN comment over the past 3 years, better use of agents and advanced models DID solve the previous temporary setbacks. The techno-optimists were right, and the nay-sayers wrong. Over the course of about 2 years, the general consensus has shifted from "it's a fun curiosity" to "it's just better stackoverflow" to "some people say it's good" to "w…

[flagged]

Your definition of a glorified autocomplete is … oof. So in short, “try ask it to do something you’d hate on bad code you’d yourself fail at and it might fail”.

And I’m pretty sure I could try Claude on a repo as you describe and it wouldn’t in fact fail. You’re letting your opinions of what LLMs were like a few months ago influence what you think of them now.

Comments like yours really annoy me because they are ridiculously confident about AI being “glorified autocomplete”, but also clearly not informed about the capabilities. I don’t get how some people can be on HN and not actually … try these things, be curious about them, try them on hard problems.

I’m a good engineer. I’ve coded for 24 years at this point. Yesterday in 45 minutes I built a feature that would have taken me three months without AI. The speed gains are obscene and because of this, we can build things we would never have even started before. Software is accelerating.

Re: Reports of code's death are greatly exaggerated

#299

Post by citrin_ru »

ehsanu1 wrote: post #189
g9yuayon wrote: post #157

Earlier quoted context omitted.

> Lattner found nothing innovative in the code generated by AI I don't think the replacement is binary. Instead, it’s a spectrum. The real concern for many software engineers is whether AI reduces demand enough to leave the field oversupplied. And that should be a question of economy: are we going to have enough new business problems to solve? If we do, AI will help us but will not replace us. If not, well, we are go…

Business problems are essentially neverending. And humans have a broader type of intelligence that LLMs lack but are needed to solve many novel problems. I wouldn't worry.

Problems are never ending but amount of money which can be made in short (or even mid) term by solving these problems is limited. Every dollar spent on LLM is a dollar not spent on salaries.

Re: Reports of code's death are greatly exaggerated

#300

Post by thesz »

fasterik wrote: post #107
thesz wrote: post #91

Earlier quoted context omitted.

> ...generate answers near the center of existing thought. This is right in the Wikipedia's article on universal approximation theorem [1]. [1] https://en.wikipedia.org/wiki/Universal_approximation_theore... "n the field of machine learning, the universal approximation theorems (UATs) state that neural networks with a certain structure can, in principle, approximate any continuous function to any desired degree of ac…

It has been proven that recurrent neural networks are Turing complete [0]. So for every computable function, there is a neural network that computes it. That doesn't say anything about size or efficiency, but in principle this allows neural networks to simulate a wide range of intelligent and creative behavior, including the kind of extrapolation you're talking about. [0] https://www.sciencedirect.com/science/article…

Judging from what I read, their work is subject to regular hardware constraints, such as limited stack size. Because paper describes a mapping from regular hardware circuits to the continuous circuits.

As an example, I would like to ask how to parse balanced brackets grammar (S ::= B ; B ::= | BB | (B) | [B] | {B};) with that Turing complete recurrent network and how it will deal with precision loss for relatively short inputs.

Paper also does not address training (i.e., automatic search of the processors' equations given inputs and outputs).

Post reply on HN