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

671–680 of 827 posts

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

#671

Post by bytefish »

bgwalter wrote: post #183

I'd like to hear from Linux kernel developers. There is no significant software that has been written (plagiarized) by "AI". Why not ask the actual experts who deliver instead of talk? This whole thing is a religion.

Microsoft is dogfooding Copilot in their dotnet/runtime [1] and dotnet/aspnetcore [2] repositories. This is the only time I have seen a company using its own AI Tools transparently. Yes, they label it an experiment, but I am pretty sure it’s mandated use within Microsoft.

I am an “AI skeptic”, so clearly I am biased here. What I am seeing in the repositories is, that Copilot hasn’t made any substantial contributions so far. The PRs, that went through? They often contain very, very detailed feedback, up to the point line by line replacements have been suggested.

The same engineers, that went up stage at “Microsoft Build 2025” to tell how amazing Copilot is and how it made them a 100x developer? They are not using Copilot in any of their PRs.

You said it’s a religion. I’d say it’s a cult. Whatever it is, outside the distortion bubble, this whole thing looks pretty bad to me.

[1] https://github.com/dotnet/runtime/pulls

[2] https://github.com/dotnet/aspnetcore/pulls

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

#672

Post by throwdbaaway »

diggan wrote: post #406
throwdbaaway wrote: post #342

Earlier quoted context omitted.

o3-mini-2025-01-31 with high reasoning effort replied with "Hi" after 448 reasoning tokens. gpt-4.5-preview-2025-02-27 replied with "Hi!"

> o3-mini-2025-01-31 with high reasoning effort replied with "Hi" after 448 reasoning tokens. I got "hi", as expected. What is the full system prompt + user message you're using? https://i.imgur.com/Y923KXB.png > gpt-4.5-preview-2025-02-27 Same "hi": https://i.imgur.com/VxiIrIy.png

Ah right, my bad. Somehow I thought the prompt was only:

    Say just 'hi'
while the "without any extra words or explanations" part was for the readers of your comment. Perhaps kubb also made a similar mistake.

I used empty system prompt.

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

#673

Post by practal »

lelanthran wrote: post #474
practal wrote: post #50

Great talk, thanks for putting it online so quickly. I liked the idea of making the generation / verification loop go brrr, and one way to do this is to make verification not just a human task, but a machine task, where possible. Yes, I am talking about formal verification, of course! That also goes nicely together with "keeping the AI on a tight leash". It seems to clash though with "English is the new programming l…

> Use informal English on the surface, while some of it is interpreted as a formal expression, put to work, and then reflected back in English? I think that is possible, if you have a formal language and logic that is flexible enough, and close enough to informal English. That sounds like a paradox. Formal verification can prove that constraints are held. English cannot. mapping between them necessarily requires disa…

Going from informal to formal can be done using autoformalization [1]. The real question is, how do you judge that the result is correct?

[1] Autoformalization with Large Language Models — https://papers.nips.cc/paper_files/paper/2022/hash/d0c6bc641...

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

#674

Post by swader999 »

whatarethembits wrote: post #655
cjcenizal wrote: post #58

Earlier quoted context omitted.

My friend Eric Pelz started a company called Malleable to do this very thing: https://www.linkedin.com/posts/epelz_every-piece-of-software...

I'm curious where this ends up going. Personally I think its a mistake; at least at "team" level. One of the most valuable things about a software or framework dictating how things are done is to give a group of people a common language to communicate with and enforce rules. This is why we generally prefer to use a well documented framework, rather than letting a "rockstar engineer" roll their own. Only they will und…

You bake those rules into the folders in a Claude.md file and it becomes it's guide when building or changing anything. Ubiquitous language and all that Jazz.

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

#675

Post by crsn »

OJFord wrote: post #661

I'm not sure about the 1.0/2.0/3.0 classification, but it did lead me to think about LLMs as a programming paradigm: we've had imperative & declarative, procedural & functional languages, maybe we'll come to view deterministic vs. probabilistic (LLMs) similarly. def __main__: You are a calculator. Given an input expression, you compute the result and print it to stdout, exiting 0. Should you be unable to do this, you…

This (sort of) is already a paradigm: https://en.m.wikipedia.org/wiki/Probabilistic_programming

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

#678

Post by dmitrijbelikov »

poorcedural wrote: post #607
dmitrijbelikov wrote: post #65

I think that Andrej presents “Software 3.0” as a revolution, but in essence it is a natural evolution of abstractions. Abstractions don't eliminate the need to understand the underlying layers - they just hide them until something goes wrong. Software 3.0 is a step forward in convenience. But it is not a replacement for developers with a foundation, but a tool for acceleration, amplification and scaling. If you know…

Foundational programmers form the base of where the seed can grow. In a way programmers found where our roots grow, they can not find your limits. Software 3.0 is a step into a different light, where software finds its own limits. If we know where they are rooted, we will merge their best attempts. Only because we appreciate their resultant behavior.

The software does nothing but what you tell it to do. And if you can't figure out the limits, then it's probably a personal problem that you haven't solved for yourself yet.

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

#679

Post by lodovic »

lymbo wrote: post #662
TeMPOraL wrote: post #430

Earlier quoted context omitted.

For me it's even simpler: V1.0: describing solutions to specific problems directly, precisely, for machines to execute. V2.0: giving machine examples of good and bad answers to specific problems we don't know how to describe precisely, for machine to generalize from and solve such indirectly specified problem. V3.0: telling machine what to do in plain language, for it to figure out and solve. V2 was coded in V1 style…

What are some predictions people are anticipating for V4? My Hail Mary is it’s going to be groups of machines gathering real world data, creating their own protocols or forms of language isolated to their own systems in order to optimize that particular system’s workflow and data storage.

But that means AGI is going to write itself

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

#680

Post by dheera »

OJFord wrote: post #661

I'm not sure about the 1.0/2.0/3.0 classification, but it did lead me to think about LLMs as a programming paradigm: we've had imperative & declarative, procedural & functional languages, maybe we'll come to view deterministic vs. probabilistic (LLMs) similarly. def __main__: You are a calculator. Given an input expression, you compute the result and print it to stdout, exiting 0. Should you be unable to do this, you…

    def __main__:
        You run main(). If there are issues, you edit __file__ to try to fix the errors and re-run it. You are determined, persistent, and never give up.
Post reply on HN