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

Andrej Karpathy: Software in the era of AI [video]

youtube.com

Post reply on HN
Print view View original

491–500 of 827 posts

Re: Andrej Karpathy: Software in the era of AI [video]

#491

Post by pton_xd »

kordlessagain wrote: post #415
darqis wrote: post #193

when I started coding at the age of 11 in machine code and assembly on the C64, the dream was to create software that creates software. Nowadays it's almost reality, almost because the devil is always in the details. When you're used to write code, writing code is relatively fast. You need this knowledge to debug issues with generated code. However you're now telling AI to fix the bugs in the generated code. I see it…

Kids? Think about all the domain experts, entrepreneurs, researchers, designers, and creative people who have incredible ideas but have been locked out of software development because they couldn't invest 5-10 years learning to code. A 50-year-old doctor who wants to build a specialized medical tool, a teacher who sees exactly what educational software should look like, a small business owner who knows their industry…

> Think about all the domain experts, entrepreneurs, researchers, designers, and creative people who have incredible ideas but have been locked out of software development because they couldn't invest 5-10 years learning to code.

> it's about unleashing all the pent-up innovation from people who understand problems deeply but couldn't translate that understanding into software.

This is just a fantasy. People with "incredible ideas" and "pent-up innovation" also need incredible determination and motivation to make something happen. LLMs aren't going to magically help these people gain the energy and focus needed to pursue an idea to fruition. Coding is just a detail; it's not the key ingredient all these "locked out" people were missing.

Re: Andrej Karpathy: Software in the era of AI [video]

#492

Post by mkleczek »

diggan wrote: post #469
mkleczek wrote: post #465

Earlier quoted context omitted.

> Would you say that ML isn't a successful discipline? ML is basically balancing between "formal language" (papers/algorithms) and "non-deterministic outcomes" (weights/inference) yet it seems useful in a wide range of applications Usefulness of LLMs has yet to be proven. So far there is more marketing in it than actual, real world results. Especially comparing to civil and mechanical engineering, maths, electrical e…

> Usefulness of LLMs has yet to be proven. What about ML (Machine Learning) as a whole? I kind of wrote ML instead of LLMs just to avoid this specific tangent. Are you feelings about that field the same?

> What about ML (Machine Learning) as a whole? I kind of wrote ML instead of LLMs just to avoid this specific tangent. Are you feelings about that field the same?

No - I only expressed my thoughts about using natural language for computing.

Re: Andrej Karpathy: Software in the era of AI [video]

#493

Post by s_ting765 »

blixt wrote: post #394

If we extrapolate these points about building tools for AI and letting the AI turn prompts into code I can’t help but reach the conclusion that future programming languages and their runtimes will be heavily influenced by the strengths and weaknesses of LLMs. What would the code of an application look like if it was optimized to be efficiently used by LLMs and not humans? * While LLMs do heavily tend towards expectin…

Given the plethora of programming languages that exist today, I'm not worried at all about AI taking over SWE jobs.

Re: Andrej Karpathy: Software in the era of AI [video]

#494

Post by poorcedural »

tudorizer wrote: post #405

95% terrible expression of the landscape, 5% neatly dumbed down analogies. English is a terrible language for deterministic outcomes in complex/complicated systems. Vibe coders won't understand this until they are 2 years into building the thing. LLMs have their merits and he sometimes aludes to them, although it almost feels accidental. Also, you don't spend years studying computer science to learn the language/synt…

Time is a funny calculator, measuring how an individual is behind. And in the funny circumstance that an individual is human, they look back on this comment in 3 years and wonder why humans only see themselves.

Re: Andrej Karpathy: Software in the era of AI [video]

#495

Post by AlexCoventry »

lubujackson wrote: post #479

Generally, people behind big revolutionary tech are the worst suited for understanding how it will do "in the wild". Forest for the trees and all that. Some good nuggets in this talk, specifically his concept that Software 1.0, 2.0 and 3.0 will all persist and all have unique use cases. I definitely agree with that. I disagree with his belief that "anyone can vibe code" mindset - this works to a certain level of fide…

I've seen evidence of "anyone can vibe code", but at this stage the result tends to be a 5,000-line application intricately entangled with 500,000 lines of irrelevant slop. Still, the wonder is that the bear can dance at all. That's a new thing under the sun.

Re: Andrej Karpathy: Software in the era of AI [video]

#496

Post by tudorizer »

brainless wrote: post #485
tudorizer wrote: post #405

95% terrible expression of the landscape, 5% neatly dumbed down analogies. English is a terrible language for deterministic outcomes in complex/complicated systems. Vibe coders won't understand this until they are 2 years into building the thing. LLMs have their merits and he sometimes aludes to them, although it almost feels accidental. Also, you don't spend years studying computer science to learn the language/synt…

I am not sure I got your point about English. I thought Karpathy was talking about English being the language of prompts, not output. Outputs can be English but if the goal is to compute using the output, then we need structured output (JSON, snippets of code, etc.), not English.

Entertain me in an exercise:

First, instruct a friend/colleague of how to multiply two 2 digit numbers in plain English.

Secondly (ideally with a different friend, to not contaminate tests), explain the same but using only maths formulas.

Where does the prompting process start and where does it end? Is it a one-off? Is the prompt clear enough? Do all the parties involved communicate within same domain objects?

Hopefully my example is not too contrived.

Re: Andrej Karpathy: Software in the era of AI [video]

#497

Post by TeMPOraL »

sensanaty wrote: post #334
TeMPOraL wrote: post #275

Earlier quoted context omitted.

Who needs water when all these hot takes come from sources so dense, they're about to collapse into black holes.

Is me not wanting the UI of my OS to shift with every mouse click a hot take? If me wanting to have the consistent "When I click here, X happens" behavior instead of the "I click here and I'm Feeling Lucky happens" behavior is equal to me being dense, so be it I guess.

No. But you interpreting and evaluating the demo in question as suggesting the things you described - frankly, yes. It takes a deep gravity well to miss a point this clear from this close.

It's a tech demo. It shows you it's possible to do these things live, in real time (and to back Karpathy's point about tech spread patterns, it's accessible to you and me right now). It's not saying it's a good idea - but there are obvious seeds of good ideas there. For one, it shows you a vision of an OS or software you can trivially extend yourself on the fly. "I wish it did X", bam, it does. And no one says it has to be non-deterministic each time you press some button. It can just fill what's missing and make additions permanent, fully deterministic after creation.

Re: Andrej Karpathy: Software in the era of AI [video]

#498

Post by tudorizer »

gardenhedge wrote: post #490
tudorizer wrote: post #452

Earlier quoted context omitted.

Working code or not is irelevant. Heck, even human-in-loop (Tony-in-the-Iron-Man) is not actively the point. If we're going into "it's all about" territory then it's all about: - training data - approximation of the desired outcome Neither support a good direction for the complexity of some of the system around us, most of which require dedicated language. Imagine doing calculus or quantum physics in English. Novels…

Tony is iron man, not in him

Sure, I wasn't sure how to call the robot layer. Is is "Iron Main Suit"?

Re: Andrej Karpathy: Software in the era of AI [video]

#499

Post by diggan »

ofjcihen wrote: post #487
diggan wrote: post #454

Earlier quoted context omitted.

> coding is no longer the problem. "Coding" - The art of literally using your fingers to type weird characters into a computer, was never a problem developers had. The problem has always been understanding and communication, and neither of those have been solved at this moment. If anything, they have gotten even more important, as usually humans can infer things or pick up stuff by experience, but LLMs cannot, and yo…

I’d take it a step further honestly. You need to be precise and exact but you also have to have enough domain knowledge to know when the LLM is making a huge mistake.

> you also have to have enough domain knowledge

I'm a bit 50/50 on this. Generally I agree, how are you supposed to review it otherwise? Blindly accepting whatever the LLM tells you or gives you is bound to create trouble in the future, you still need to understand and think about what the thing you're building is, and how to design/architect it.

I love making games, but I'm also terrible at math. Sometimes, I end up out of my depth, and sometimes it could take me maybe a couple of days to solve something that probably would be trivial for a lot of people. I try my best to understand the fundamentals and the theory behind it, but also not get lost in rabbit holes, but it's still hard, for whatever reason.

So I end up using LLMs sometimes to write small utility functions used in my games for specific things. It takes a couple of minutes. I know exactly what I want to pass into it, and what I want to get back, but I don't necessarily understand 100% of the math behind it. And I think I'm mostly OK with this, as long as I can verify that the expected inputs get the expected outputs, which I usually do with unit or E2E tests.

Would I blindly accept information about nuclear reactors, another topic I don't understand much about? No, I'd still take everything a LLM outputs with a "grain of probability" because that's how they work. Would I blindly accept it if I can guarantee that for my particular use case, it gives me what I expect from it? Begrudgingly, yeah, because I just wanna create games and I'm terrible at math.

Re: Andrej Karpathy: Software in the era of AI [video]

#500

Post by jbeninger »

diggan wrote: post #462
tudorizer wrote: post #456

Earlier quoted context omitted.

> Did you talk with them? Otherwise this is just creating an imaginary argument against some people you just assume they didn't listen. I have, unfortunately. Start-up founders, managers, investors who taunt the need for engineers because "AI can fix it". Don't get me wrong, there are plenty of "stochastic parrot" engineers even without AI, but still, not enough to make blanket statements.

That's a lot of people to talk to in a day more or less, since the talk happened. Were they all there and you too, or you all had a watch party or something? Still, what's the outcome of our "glazed donut" argument, you got me curious what that would lead to. Did I die of diabetes?

I think the analogy is that vibe coding is bad for you but feels good. Like a donut.

But I'd say the real situation is more akin to "if you eat this donut quickly, you might get diabetes, but if you eat it slowly, it's fine", which is a bad analogy, but a bit more accurate.

Post reply on HN