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

Don't fall into the anti-AI hype

antirez.com

Post reply on HN
Print view View original

721–730 of 1001 posts

Re: Don't fall into the anti-AI hype

#721

Post by XenophileJKO »

unyttigfjelltol wrote: post #604
totallykvothe wrote: post #569

I don't understand the stance that AI currently is able to automate away non-trivial coding tasks. I've tried this consistently since GPT 3.5 came out, with every single SOTA model up to GPT 5.1 Codex Max and Opus 4.5. Every single time, I get something that works, yes, but then when I start self-reviewing the code, preparing to submit it to coworkers, I end up rewriting about 70% of the thing. So many important deta…

> non-trivial coding tasks I’ve come back to the idea LLMs are super search engines. If you ask it a narrow, specific question, with one answer, you may well get the answer. For the “non-trivial” questions, there always will be multiple answers, and you’ll get from the LLM all of these depending on the precise words you use to prompt it. You won’t get the best answer, and in a complex scenario requiring highly recurs…

I'm not going to argue about how capable the models are, I personally think they are pretty capable.

What I will argue is that the LLMs are not just search engines. They have "compressed" knowledge. When they do this, they learn relations between all kinds of different levels of abstractions and meta patterns.

It is really important to understand that the model can follow logical rules and has some map of meta relationships between concepts.

Thinking of a LLM as a "search engine" is just fundamentally wrong in how they work, especially when connected to external context like code bases or live information.

Re: Don't fall into the anti-AI hype

#722

Post by PeterStuer »

unyttigfjelltol wrote: post #604
totallykvothe wrote: post #569

I don't understand the stance that AI currently is able to automate away non-trivial coding tasks. I've tried this consistently since GPT 3.5 came out, with every single SOTA model up to GPT 5.1 Codex Max and Opus 4.5. Every single time, I get something that works, yes, but then when I start self-reviewing the code, preparing to submit it to coworkers, I end up rewriting about 70% of the thing. So many important deta…

> non-trivial coding tasks I’ve come back to the idea LLMs are super search engines. If you ask it a narrow, specific question, with one answer, you may well get the answer. For the “non-trivial” questions, there always will be multiple answers, and you’ll get from the LLM all of these depending on the precise words you use to prompt it. You won’t get the best answer, and in a complex scenario requiring highly recurs…

An example I had last month. Some code (dealing with PDF's) package ran into a resources problem in production. LLM suggested an adaptation to the segment that caused the problem, but that code pulled in 3 new non-trivial dependecies. Added constraints and the next iteration it dropped 1 of the 3. Pushed further and it confirmed my suggestion that the 2 remaining dependencies could be covered just by specifying an already existing parameter in the constructor.

The real problem btw was a bug introduced in the PDF handeling package 2 versions ago that caused resource handeling problems in some contexts, and the real solution was roling back to the version before the bug.

I'm still using AI daily in my development though, as as long as you sort of know what you are doing and have enough knowledge to evaluate it is very much a net productivity multiplier for me.

Re: Don't fall into the anti-AI hype

#723

Post by Madmallard »

saxenaabhi wrote: post #619
totallykvothe wrote: post #569

I don't understand the stance that AI currently is able to automate away non-trivial coding tasks. I've tried this consistently since GPT 3.5 came out, with every single SOTA model up to GPT 5.1 Codex Max and Opus 4.5. Every single time, I get something that works, yes, but then when I start self-reviewing the code, preparing to submit it to coworkers, I end up rewriting about 70% of the thing. So many important deta…

Why not post a github gist with prompt and code so that people here can give you their opinion?

Those just don't appear at all on HackerNews

Gee I wonder why

Re: Don't fall into the anti-AI hype

#725

Post by rbbydotdev »

totallykvothe wrote: post #569

I don't understand the stance that AI currently is able to automate away non-trivial coding tasks. I've tried this consistently since GPT 3.5 came out, with every single SOTA model up to GPT 5.1 Codex Max and Opus 4.5. Every single time, I get something that works, yes, but then when I start self-reviewing the code, preparing to submit it to coworkers, I end up rewriting about 70% of the thing. So many important deta…

> I end up rewriting about 70% of the thing.

I think this touches on the root of the issue. I am seeing a results over process winning. Code quality will reduce. Out of touch or apathetic project management who prioritize results, now are even more emboldened to have more tech debt riddled code

Re: Don't fall into the anti-AI hype

#726

Post by reducesuffering »

dist-epoch wrote: post #83
cmiles8 wrote: post #6

The “anti-AU hype” phrase oversimplifies what’s playing out at the moment. On the tech side, while things are a bit rough around the edges still the tech is very useful and isn’t going away. I honestly don’t see much disagreement there. The concern mostly comes from the business side… that for all the usefulness on the tech there is no clearly viable path that financially supports everything that’s going on. It’s a n…

People said the exact same thing about (numbers from memory, might be off): - when Google paid $1 bil for YouTube - when Facebook paid $1 bil for Instagram - when Facebook paid $1 bil for WhatsApp The same thing - these 3 companies make no money, and have no path to making money, and that the price paid was crazy and decoupled from any economics. Yet now, in hindsight, they look like brilliant business decisions.

They were even saying this about Uber just a couple years ago. Now Uber makes $15b a year

Re: Don't fall into the anti-AI hype

#727

Post by carlmr »

unyttigfjelltol wrote: post #604
totallykvothe wrote: post #569

I don't understand the stance that AI currently is able to automate away non-trivial coding tasks. I've tried this consistently since GPT 3.5 came out, with every single SOTA model up to GPT 5.1 Codex Max and Opus 4.5. Every single time, I get something that works, yes, but then when I start self-reviewing the code, preparing to submit it to coworkers, I end up rewriting about 70% of the thing. So many important deta…

> non-trivial coding tasks I’ve come back to the idea LLMs are super search engines. If you ask it a narrow, specific question, with one answer, you may well get the answer. For the “non-trivial” questions, there always will be multiple answers, and you’ll get from the LLM all of these depending on the precise words you use to prompt it. You won’t get the best answer, and in a complex scenario requiring highly recurs…

>It’s not readily apparent at first blush the LLM is doing this, giving all the answers.

Now I'm wondering if I'm prompting wrong. I usually get one answer. Maybe a few options but rarely the whole picture.

I do like the super search engine view though. I often know what I want, but e.g. work with a language or library I'm not super familiar with. So then I ask how do I do x in this setting. It's really great for getting an initial idea here.

Then it gives me maybe one or two options, but they're verbose or add unneeded complexity. Then I start probing asking if this could be done another way, or if there's a simpler solution to this.

Then I ask what are the trade-offs between solutions. Etc.

It's maybe a mix of search engine and rubber ducking.

Agents are, like for OP, a complete failure for me though. Still can't get them to not run off into a completely strange direction, leaving a minefield of subtle coding errors and spaghetti behind.

Re: Don't fall into the anti-AI hype

#728

Post by matwood »

fmbb wrote: post #703
khafra wrote: post #667

Earlier quoted context omitted.

> Non-trivial coding tasks A coding agent just beat every human in the AtCoder Heuristic optimization contest. It also beat the solution that the production team for the contest put together. https://sakana.ai/ahc058/ It's not enterprise-grade software, but it's not a CRUD app with thousands of examples in github, either.

> It's not enterprise-grade software, but it's not a CRUD app with thousands of examples in github, either. Optimization is a very simple problem though. Maintaining a random CRUD app from some startup is harder work.

> Optimization is a very simple problem though.

C'mon, there's post every other week that optimization never happens anymore because it's too hard. If AI can take all the crap code humans are writing and make it better, that sounds like a huge win.

Re: Don't fall into the anti-AI hype

#729

Post by ahtihn »

Closi wrote: post #713
totallykvothe wrote: post #569

I don't understand the stance that AI currently is able to automate away non-trivial coding tasks. I've tried this consistently since GPT 3.5 came out, with every single SOTA model up to GPT 5.1 Codex Max and Opus 4.5. Every single time, I get something that works, yes, but then when I start self-reviewing the code, preparing to submit it to coworkers, I end up rewriting about 70% of the thing. So many important deta…

Try learning to vibe code on something totally greenfield without looking at the code and see if it changes your mind. Ignore code quality, “does it work” and “am i happy with the app” are the only metrics. Code quality is an issue you need to ignore with vibe coding - if code quality is important to your project or you then it’s not an issue. But if you abandon this concept and build things small enough or modular e…

If you start greenfield and ignore the code quality, how do you know you can maintain it long term?

Greenfield is fundamentally easier than maintaining existing software. Once software exists, users expect it to behave a certain way and they expect their data to remain usable in new versions.

The existing software now imposes all sorts of contraints that may not be explicit in the spec. Some of these constraints end up making some changes very hard. Bad assumptions in data modeling can make migrations a nightmare.

You can't just write entirely new software every time the requirements change.

Re: Don't fall into the anti-AI hype

#730

Post by matwood »

sublinear wrote: post #635
daxfohl wrote: post #599

Earlier quoted context omitted.

Agreed, but: There's been a notable jump over the course of the last few months, to where I'd say it's inevitable. For a while I was holding out for them to hit a ceiling where we'd look back and laugh at the idea they'd ever replace human coders. Now, it seems much more like a matter of time. Ultimately I think over the next two years or so, Anthropic and OpenAI will evolve their product from "coding assistant" to "…

This doesn't make any sense. If the business can get rid of their engineers, then why can't the user get rid of the business providing the software? Why can't the user use AI to write it themselves? I think instead the value is in getting a computer to execute domain-specific knowledge organized in a way that makes sense for the business, and in the context of those private computing resources. It's not about the abi…

> If the business can get rid of their engineers, then why can't the user get rid of the business providing the software?

I have't checked the stats lately, but at one point most software written was in non-tech companies for the single business. The first 1/2 of my career was spent writing in-house software for a company that did everything from custom reporting and performance tracking to scraping data of automated phone dialers. There's so much software out there that effectively has a user base of a single company.

Post reply on HN