Agent Skills by sunxxuns
Showing 3 vetted skills indexed across 1 GitHub repositories.
sunxxunsbenchmark-kernel
Benchmark FlashInfer GPU kernel performance with CUPTI timing across backends.
Community
Advanced
sunxxunsdebug-cuda-crash
Diagnose CUDA crashes by analyzing FlashInfer API logs in PyTorch workloads.
Community
Advanced
sunxxunsadd-cuda-kernel
Automate adding a CUDA kernel to FlashInfer with launcher, TVM-FFI binding, JIT generator, Python API, tests, and AOT registration.
Community
Advanced