Learn CUDA × LLM inference from scratch: 14-chapter Chinese tutorial + runnable CUDA C++ code, from Hello CUDA to a hand-written GPT-2 inference engine
1 file for Claude Code: learn-cuda-from-scratch CLAUDE.md — 1,557 tokens loaded in every session.
CLAUDE.md A 1,557 tok These files are jwzheng96/learn-cuda-from-scratch's own configuration — they tell Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.