← all publishers

troycheng

@troycheng source repo

1 published skill

  1. Cuda Kernel Optimizer · troycheng bundle
    Use when optimizing or diagnosing GPU workload performance with CUDA, CUTLASS, Triton, PyTorch, vLLM or TensorRT-LLM; interpreting NCU, Nsys or PyTorch Profiler evidence; or preparing the workload and correctness checks for such optimization. General coding, skill maintenance and conceptual GPU questions do not require the optimization workflow.
    0
    installs