NVIDIA/TileGym/converting-cutile-to-julia A
NVIDIA/skills
Agent Skill
Active
Converts cuTile Python GPU kernels (@ct.kernel) to cuTile.jl Julia equivalents.
Add a new cuTile GPU kernel operator to TileGym.
Install this skill (Claude Code)
# Clone and copy the skill into your project
git clone https://github.com/NVIDIA/skills.git
mkdir -p .claude/skills
cp -r skills/skills/TileGym/adding-cutile-kernel .claude/skills/
# Or for personal use: ~/.claude/skills/Transparent heuristic — same formula for every entry. Total 85/100.
Converts cuTile Python GPU kernels (@ct.kernel) to cuTile.jl Julia equivalents.
Converts cuTile GPU kernels (@ct.kernel) to Triton (@triton.jit).
Use when adding, modifying, optimizing, or debugging CuTile autotuning code.
Expert cuTile programming assistant.
Iteratively optimize cuTile kernel performance through systematic profiling, bottleneck analysis, IR comparison, and targeted tuning.
Integrate TileGym kernels into Hugging Face `transformers` models by replacing the library's submodule(s) and certain class(es)' implementations, and patching certain class(es)' in...