Hacker News Reader

Comment by augment_me | Hacker News Reader

I am with you on this, and you can't win, because as soon as you voice this opinion you get overwhelmed with "you dont have the sauce/prompt" opinions which hold an inherent fallacy because they assume you are solving the same problems as them.

I work in GPU programming, so there is no way in hell that JavaScript tools and database wrapper tasks can be on equal terms with generating for example Blackwell tcgen05 warp-scheduled kernels.

View on HN

There's going to be a long tail of domain-specific tasks that aren't well served by current models for the foreseeable future, but there's also no question the complexity horizon of the SotA models is increasing over time. I've had decent results recently with non-trivial Cuda/MPS code. Is it great code/finely tuned? Probably not but it delivered on the spec and runs fast enough.

> you can't win, because as soon as you voice this opinion you get overwhelmed with "you dont have the sauce/prompt"

> I've had decent results recently with non-trivial Cuda/MPS code.

Anthropic has a challenge of optimizing GPU code.

The current leader is Opus 4.5

https://github.com/anthropics/original_performance_takehome

I have done it, its not GPU-code, you are optimizing a toy compiler for a fictional framework. There is some SIMD mechanics but you cant call it GPU. There is a lot of such real challenges though - KernelBench, Project Popcorn, FlashInfer, Wafer, Standard Kernel.

In my experience LLMs are useless for GPU compute code, just not enough in the training set.

Yeah, the argument here is that once you say this, people will say "you just dont know how to prompt, i pass the PTX docs together with NSight output and my kernel into my agent and run an evaluation harness and beat cuBLAS". And then it turns out that they are making a GEMM on Ampere/Hopper which is an in-distribution problem for the LLMs.

It's the idea/mindset that since you are working on something where the tool has a good distribution, its a skill issue or mindset problem for everyone else who is not getting value from the tool.

Now please get back to coding GPU stuff so we can train our models on your code. Thank you.

Another thing I've never got them to generate is any G code. Maybe that'll be in the image/3d generator side indirectly, but I was kind of hoping I could generate some motions since hand coding coordinates is very tedious. That would be a productivity boost for me. A very very niche boost, since I rarely need bespoke G code, but still.

Oh HELL no. :P Gcode is (at least if you’re talking about machining) the very definition of something you want to generate analytically using tried and tested algorithms with full consideration taken for the specifics of the machine and material involved.

I guess if you just want to use it to wiggle something around using a stepper motor and a spare 3D printer control board, it might be OK though. :)

Reply on news.ycombinator.com