Skip to content
agentusehub
Sign in+ Submit
Software & DevOpsAny

Optimize CUDA kernels with agents that benchmark and profile

AI agents run a C++ CUDA test harness, get benchmarks, and profile with Nsight to optimize kernels, built with LangGraph.

Read the experience
Time saved
Not reported
Setup time
Not reported
Running cost
Not reported
Cost range
Not reported

Use case overview

AI agents run a C++ CUDA test harness, get benchmarks, and profile with Nsight to optimize kernels, built with LangGraph.

Read the original report ↗

Reported experience

  1. What they did

    The user gives a workload description, or supplies their own signature, reference, kernel and input cases. A LangGraph agent proposes kernel and launch-config changes. A C++ harness compiles them with NVRTC and runs them through the CUDA Driver API. Python checks outputs against NumPy and times each candidate. Failed candidates get repaired within the iteration budget. Optional flags add Nsight Compute profiling and NVIDIA documentation research.

  2. How it went

    The README shows float32 GEMM runs on an RTX 3060 Laptop GPU, with a second run continuing from the first run's best kernel. It states no speedup numbers, and no comparison against cuBLAS.

  3. Worth knowing

    Generated input scripts run unsandboxed on your machine and API usage (default gpt-5-mini) is billed to you. A generated reference is not an independent check of correctness.

Example prompts

Try with Claude Code
Optimize the CUDA kernel in [file path] for [operation, e.g. matrix multiply]. Build a test harness that checks correctness, benchmark the baseline, profile it with Nsight, and try improvements one at a time, keeping only those that pass the correctness check. Stop when you have a faster version and a table of each change with its speedup.

Related use cases

Software & DevOps

Port a Chrome extension to Safari

The author had ChatGPT take the Session Buddy extension source from Chrome, port it to Safari and install it, taking about 20 minutes.

ChatGPT