aa24aa

cuda-optimized-skill. A CUDA kernel optimization toolkit for validation, benchmarking, Nsight Compute profiling, bottleneck analysis, and iterative tuning. It helps improve custom GPU operators with reproducible workflows and evidence-based performance comparison.

cuda-opt-agent. CUDA Opt Agent is an LLM-driven toolkit for automated CUDA kernel optimization. It generates baselines from text, specs, or existing .cu files, compiles and validates with nvcc, benchmarks on real GPUs, profiles with Nsight Compute, iterates improvements, and preserves reusable tuning knowledge.