- Email: chenyucang223@gmail.com
I'm an undergraduate student exploring AI Infra, GPU Programming, and LLM Inference Optimization.
Currently learning and building around:
- CUDA kernels: Softmax, RMSNorm, GEMM, FlashAttention
- LLM inference: KV Cache, continuous batching, vLLM
- GPU programming frameworks: CUDA and TileLang
- C++ systems programming
An AI-assisted pull request review tool that fetches GitHub PR context, combines static analysis signals with an AI/mock model, and generates structured review findings.
A learning repo for implementing and optimizing common deep learning kernels such as Softmax, RMSNorm, and GEMM.
A C++17 HTTP/1.1 server project for practicing Linux networking, epoll, non-blocking I/O, thread pools, and RAII.

