Earlier quoted context omitted.
[flagged]
" Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something. " https://news.ycombinator.com/newsguidelines.html
Fair
Andrej Karpathy: Software in the era of AI [video]
151–160 of 827 posts
Re: Andrej Karpathy: Software in the era of AI [video]
#152Re: Andrej Karpathy: Software in the era of AI [video]
#153It’s fascinating to think about what true GUI for LLM could be like. It immediately makes me think a LLM that can generate a customized GUI for the topic at hand where you can interact with in a non-linear way.
This talk https://www.youtube.com/watch?v=MbWgRuM-7X8 explores the idea of generative / malleable personal user interfaces where LLMs can serve as the gateway to program how we want our UI to be rendered.
Re: Andrej Karpathy: Software in the era of AI [video]
#154He sounds like Terrence Howard with his nonsense.
Re: Andrej Karpathy: Software in the era of AI [video]
#155You can generate 1.0 programs with 3.0 programs. But can you generate 2.0 programs the same way?
2.0 programs (model weights) are created by running 1.0 programs (training runs). I don't think it's currently possible to ask a model to generate the weights for a model.
But you can generate synthetic data using a 3.0 program to train a smaller, faster, cheaper-to-run 2.0 program.
Re: Andrej Karpathy: Software in the era of AI [video]
#156The beginning was painful to watch as is the cheering in this comment section. The 1.0, 2.0, and 3.0 simply aren't making sense. They imply a kind of a succession and replacement and demonstrate a lack of how programming works. It sounds as marketing oriented as "Web 3.0" that has been born inside an echo chamber. And yet halfway through, the need for determinism/validation is now being reinvented. The analogies make…
The whole AI scene is starting to feel a lot like the cryptocurrency bubble before it burst. Don’t get me wrong, there’s real value in the field, but the hype, the influencers, and the flashy “salon tricks” are starting to drown out meaningful ML research (like Apple's critical research that actually improves AI robustness). It’s frustrating to see solid work being sidelined or even mocked in favor of vibe-coding.
Meanwhile, I asked this morning Claude 4 to write a simple EXIF normalizer. After two rounds of prompting it to double-check its code, I still had to point out that it makes no sense to load the entire image for re-orientating if the EXIF orientation is fine in the first place.
Vibe vs reality, and anyone actually working in the space daily can attest how brittle these systems are.
Re: Andrej Karpathy: Software in the era of AI [video]
#157Meanwhile, I asked this morning Claude 4 to write a simple EXIF normalizer. After two rounds of prompting it to double-check its code, I still had to point out that it makes no sense to load the entire image for re-orientating if the EXIF orientation is fine in the first place.
Vibe vs reality, and anyone actually working in the space daily can attest how brittle these systems are.
Maybe this changes in SWE with more automated tests in verifiable simulators, but the real world is far to complex to simulate in its vastness.
Re: Andrej Karpathy: Software in the era of AI [video]
#158Earlier quoted context omitted.
The safety part will probably be either solved or a non-issue or ignored. Similarly to how GPT3 was often seen as dangerous before ChatGPT was released. Some people who have only ever vibe coded are finding jobs today, ignoring safety entirely and lacking a notion of it or what it means. They just copy paste output from ChatGPT or an agentic IDE. To me it's JIT already with extra steps. Or they have pivoted their sof…
As "jit" to you means running code, and not "building and executing machine code", maybe you could vibe code this. And enjoy the segfaults.
In a way he's making sense. If the "code" is the prompt, the output of the llm is an intermediate artifact, like the intermediate steps of gcc.
So why should we still need gcc?
The answer is of course, that we need it because llm's output is shit 90% of the time and debugging assembly or binary directly is even harder, so putting asides the difficulties of training the model, the output would be unusable.
Re: Andrej Karpathy: Software in the era of AI [video]
#159There were some cool ideas- I particularly liked "psychology of AI" Overall though I really feel like he is selling the idea that we are going to have to pay large corporations to be able to write code. Which is... terrifying. Also, as a lazy developer who is always trying to make AI do my job for me, it still kind of sucks, and its not clear that it will make my life easier any time soon.
He says that now we are in the mainframe phase. We will hit the personal computing phase hopefully soon. He says llama (and DeepSeek?) are like Linux in a way, OpenAI and Claude are like Windows and MacOS.
So, No, he’s actually saying it may be everywhere for cheap soon.
I find the talk to be refreshingly intellectually honest and unbiased. Like the opposite of a cringey LinkedIn post on AI.
Re: Andrej Karpathy: Software in the era of AI [video]
#160Tight feedback loops are the key in working productively with software. I see that in codebases up to 700k lines of code (legacy 30yo 4GL ERP systems). The best part is that AI-driven systems are fine with running even more tight loops than what a sane human would tolerate. Eg. running full linting, testing and E2E/simulation suite after any minor change. Or generating 4 versions of PR for the same task so that the h…
I don't think the human is the problem here, but the time it takes to run the full testing suite.