Show HN: Agentic CUDA Kernel Optimizer

Show HN: Agentic CUDA Kernel Optimizer github.com

A langgraph based workflow with a C++ CUDA harness to optimize CUDA kernels - bertaye/agentic-cuda-optimizer

37 points•bertaye•6 days ago•11 comments•

#cuda #optimizer #agentic #agentic cuda #kernel #cuda kernel #kernel optimizer #agentic cuda kernel

Hello; I was working on optimizing some CUDA kernels and I thought may be it is a good oppurtunity learn langgraph as well. I created a simple C++ CUDA Test Harness and handed that to AI agents. They can run kernels, get benchmarks, and even can profile via nsight

11 comments

fooblaster6 days ago

Can someone explain why this isn't better accomplished through a single prompt to Claude code or codex? I don't think I understand.

bertaye5 days ago

Hello, indeed you can just use that.

The basic idea here is just automating and limiting the steps that AI can take. These are described as ‘nodes’ and their actions are limited/more descriptive from developer perspective.

The langgraph simply allows you to set some fences around the AI agent for a goal, instead of raw terminal flow. Is it better? Arguable.

osti6 days ago

Yup, this thing is basically useless. I just did /goal optimize the cuda kernel, and ai agent just proposed and tested a bunch of ideas by itself, and it profiled them using nsight ncu etc. by itself, which lead to one order of magnitude faster kernel.

generalizations6 days ago

Very cool. Did you also try using the karpathy autoresearch? How do you think this compares?

bertaye6 days ago

honestly I know it exists but I never used it so can't compare

aidiveyt5 days ago

in mine, only an agent's final report reaches the orchestrator, never its transcript, so a wrong turn inside a node stays invisible. that's the fence i'd want first.

Read the full thread on Hacker News →

11 comments

Related stories