HPC-AI-Optimization-LabCUDA kernel optimization lab

Explore GEMM, FlashAttention, quantization, and GPU performance learning from one focused repository.

📚

Progressive learning

Follow the documentation from foundational memory patterns to more advanced GPU kernels and architecture-specific notes.

🧭

Focused project surface

Start from the guides, examples, and architecture notes without having to dig through a duplicated docs surface.

🧪

Code plus context

Use the guides, examples, architecture notes, and API pages together instead of treating the site as a README mirror.

What this project is for

learning how CUDA kernel optimization techniques build on one another
reading a compact codebase that covers multiple AI/HPC kernel categories
understanding how the repository is organized, documented, and validated

Recommended starting points

Goal	Start here
Set up and build the project	/en/guide/installation
Run a first example	/en/guide/quick-start
Study optimization patterns	/en/guide/memory
Jump straight to matrix multiplication	/en/guide/gemm
See architecture notes	/en/api/architecture

Why this docs site exists

This site is a compact landing layer for the repository. Use it to choose a learning path, jump into examples, and find the right architecture or API reference page without treating the site as a README mirror.