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

How I program with LLMs

crawshaw.io

Post reply on HN
Print view View original

331–340 of 342 posts

Re: How I program with LLMs

#331

Post by owebmaster »

lm28469 wrote: post #189
owebmaster wrote: post #180

I thought his project, sketch.dev is of very poor quality. I wouldn't ship something like this - the auth process is awful and broke, I still can't login. If after 14 hours of the post the service is still rugged to death, it also means the scalability of the app is bad. If we are going to use LLMs to replace hours of programming, we should aim for quality too.

It's really bad, much less useful than even the first public version of chatgpt. Even once you manage to log in, most of the time it doesn't even give something that compiles, it calls functions/variables which don't exist. The first line of the main had 2 errors...

I finally could login and I would prefer I had not. Honestly, it is pathetic that something like this has so many positive comments. I guess most people commenting didn't even try sketch.dev

Re: How I program with LLMs

#332

Post by IanKerr »

namaria wrote: post #322
liotier wrote: post #132

Earlier quoted context omitted.

> If you are good at doing something, you might find the new tool's output to be sub-par over what you can achieve yourself, but often the lower quality output comes much faster than you can generate. That causes the people who are deliberate & precise about their process to hate the new tool completely Wow, I've been there ! Years ago we dragged a GIS system kicking and screaming from its nascent era of a dozen ultr…

This whole thing makes me think of that short story "The Machine Stops". As we keep burrowing deeper and deeper into an overly complex system that allows people to get into parts of it without understanding the whole, we are edging closer to a situation where no one is left who can actually reason about the system and it starts to deteriorate beyond repair until it suddenly collapses.

We are so, so far beyond that point already. The complexity of the world economy is beyond any one mind to fully comprehend. The microcosm of building black-box LLMs that perform feats we don't understand is yet another instance of us building systems which may forever be beyond human understanding.

How is any human meant to understand a billion lines of code in a single codebase? How is any human meant to understand a world where there are potentially trillions of lines of code operating?

Re: How I program with LLMs

#333

Post by cudgy »

ThePyCoder wrote: post #328
cudgy wrote: post #206

Earlier quoted context omitted.

So why bother securing anything at all if not willing to secure the raisons d'être? Doesn’t that suggest that these companies are trivial entities?

Only if you see source code as the only valuable thing, which it isn't. The knowledge of the team, industry connections, experience etc etc are a big part of what make it so you can effectively use the source code. We're making an industrial sorting machine. Our management is feared to death to lose the source code. But realistically, who's going to put in the time to fully understand a codebase we can barely grasp o…

You just supported my point that these companies at their core have little value. A team? Teams are fleeting and easily replaced given the hiring and firing (and poaching) practices of companies. Industry connections? Maybe to some degree, but those are fleeting as well and how do you value it? Most of these connections are held by relatively few people in the company.

Companies in other legal jurisdictions will and can steal ip with little impunity and throw new AI tools to quickly gather an understanding of the codebase. Furthermore, knowledge of source provides a roadmap to attack vectors for security violations. Seems foolish to dismiss the risks of losing control of source code.

Re: How I program with LLMs

#334

Post by namaria »

IanKerr wrote: post #332
namaria wrote: post #322

Earlier quoted context omitted.

This whole thing makes me think of that short story "The Machine Stops". As we keep burrowing deeper and deeper into an overly complex system that allows people to get into parts of it without understanding the whole, we are edging closer to a situation where no one is left who can actually reason about the system and it starts to deteriorate beyond repair until it suddenly collapses.

We are so, so far beyond that point already. The complexity of the world economy is beyond any one mind to fully comprehend. The microcosm of building black-box LLMs that perform feats we don't understand is yet another instance of us building systems which may forever be beyond human understanding. How is any human meant to understand a billion lines of code in a single codebase? How is any human meant to understand…

When your house is on fire and someone says "get out", certainly grabbing a jerrycan of gasoline and dousing yourself in fuel is worst than just getting out?

Re: How I program with LLMs

#335

Post by alightsoul »

dns_snek wrote: post #214
harrall wrote: post #207

Earlier quoted context omitted.

I believe it’s more that people hate trying new tools because they’ve already made their choice and made it their identity. However, there are also people who love everything new and jump onto the latest hype too. They try new things but then immediately advocate it without merit. Where are the sane people in the middle?

As an experienced software developer, I paid for ChatGPT for a couple of months, I trialed Gemini Pro for a couple of months, and I've used the current version of Claude. I'd be happy if LLMs could produce working code as often and as quickly as the evangelist claim, but whenever I try to use LLM to work on my day to day tasks, I almost always walk away frustrated and disappointed - and most of my work is boring on t…

I have, in person. Their code requires a lot of cleanup and then there's the AI pit of death they often are not able to crawl out of due to... Mostly language differences. They don't know enough English to look stuff up and figure out how to fix things. Programming resources in other languages are pretty much non-existent

Re: How I program with LLMs

#336

Post by alightsoul »

codebje wrote: post #10
notjoemama wrote: post #7

Our company has a no AI use policy. The assumption is zero trust. We simply can’t know whether a model or its framework could or would send proprietary code outside the network. So it’s best to assume all LLMs/AI is or will send code or fragments of code. While I applaud the incredible work by their creators, I’m not sure how a responsible enterprise class company could rely on “trust us bro” EULAs or repo readmes.

The same way responsible enterprise class companies rely on "trust us bro" EULAs for financial systems, customer databases, payroll, and all the other systems it would be very expensive and error prone to build custom for every business.

But isn't AI very competitive right now? And also have more direct access to the "secret sauce" of the company?

Re: How I program with LLMs

#337

Post by alightsoul »

ryanobjc wrote: post #280
codebje wrote: post #10

Earlier quoted context omitted.

The same way responsible enterprise class companies rely on "trust us bro" EULAs for financial systems, customer databases, payroll, and all the other systems it would be very expensive and error prone to build custom for every business.

Pretty much this. OpenAI poisoned the well badly with their "we train off your chats" nonsense. If you are using any API service, or any enterprise ChatGPT plan, your tokens are not being logged and recycled into new training data. As for why trust them? Like the parent said: EULAs. Large companies trust EULAs and terms of service for every single SAAS product they use, and they use tons and tons of them. OpenAI in a…

The enterprise plan and I believe the API too are more expensive

Re: How I program with LLMs

#338

Post by alightsoul »

attentive wrote: post #37
notjoemama wrote: post #7

Our company has a no AI use policy. The assumption is zero trust. We simply can’t know whether a model or its framework could or would send proprietary code outside the network. So it’s best to assume all LLMs/AI is or will send code or fragments of code. While I applaud the incredible work by their creators, I’m not sure how a responsible enterprise class company could rely on “trust us bro” EULAs or repo readmes.

So, you're asking how enterprise class companies are using github for repos and gmail for all the enterprise mail? What's next, zoom/teams for meetings?

They might be using neither

Re: How I program with LLMs

#339

Post by rudiksz »

HappMacDonald wrote: post #275
rudiksz wrote: post #244

Earlier quoted context omitted.

> "seasoned programmers are using LLMs better". I do not remember a single instance when code provided to me by an LLM worked at all. Even if I ask something small that cand be done in 4-5 lines of code is always broken. From a fellow "seasoned" programmer to another: how the hell do you write the prompts to get back correct working code?

I'd ask things like "which LLM are you using", and "what language or APIs are you asking it to write for". For the standard answers of "GPT-4 or above", "claude sonnet or haiku", or models of similar power and well known languages like Python, Javascript, Java, or C and assuming no particularly niche or unheard of APIs or project contexts the failure rate of 4-5 line of code scripts in my experience is less than 1%.

It's mostly Go, some Python, and I'm not asking anything niche. I'm asking for basic utility functions that I could implement in 10-20 lines of code. There's something broken every single time and I spend more time debugging the generated code than actually writing it out.

I'm pretty sure everybody measures "failure rate" differently and grossly exaggerate the success rate. There's a lot of suggestions below about "tweaking", but if I have to "tweak" generated code in any way then that is a failure for me. So the failure rate of generated code is about 99%.

Re: How I program with LLMs

#340

Post by ghostbit »

> you’re going to have days of tense back-and-forth about whether the cost of the work is worth the benefit. An LLM will do it in 60 seconds and not make you fight to get it done. Take advantage of the fact that redoing work is extremely cheap.

The fast iteration cycle of getting a baseline (but less than ideal or even completely wrong) is a great point here. Redoing the work is fast and easy but still requires review and validation to know how to request the rework to obtain the optimal result.

Post reply on HN