Skip to content
View Hayden727's full-sized avatar
🎯
Focusing
🎯
Focusing
9 followers · 51 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: Quickdraw

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: Quickdraw

Block or report Hayden727

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Hayden727/README.md

Hi, I'm Chenchen Hong 👋

AI Infrastructure · MLSys · Compilers

I build and optimize the systems that make large models fast — from compiler-level kernel work up to distributed inference serving.

🚀 What I work on

  • Multimodal & LLM Inference Infrastructure (main focus) — performance engineering for multimodal serving on SGLang-omni, alongside LLM serving stacks (SGLang, vLLM): model integration, scheduling, memory efficiency, and throughput/latency optimization.
  • RL Infrastructure — systems and tooling for reinforcement learning workloads: training/inference orchestration, rollout, and scaling.
  • Kernel Compiler Optimization — compiler-driven kernel optimization for ML workloads: codegen, graph-level transformations, and automatic kernel generation/tuning (Triton, CUDA) on NVIDIA Hopper (H100) and Blackwell (B200).

🛠️ Tech & Tools

Python C++ CUDA Triton PyTorch

📊 GitHub Stats

GitHub stats Top languages

GitHub streak

📫 Reach me

Weekly Issue Arena

Pinned Loading

  1. sgl-project/sglang-omni sgl-project/sglang-omni Public

    SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models

    Python 522 217

  2. sgl-project/sglang sgl-project/sglang Public

    SGLang is a high-performance serving framework for large language models and multimodal models.

    Python 29.5k 6.6k

  3. sgl-project/SpecForge sgl-project/SpecForge Public

    Train speculative decoding models effortlessly and port them smoothly to SGLang serving.

    Python 900 262

  4. NVIDIA-NeMo/Automodel NVIDIA-NeMo/Automodel Public

    🚀 Pytorch Distributed native training library for LLMs/VLMs with OOTB Hugging Face support

    Python 610 185

  5. ctorch ctorch Public

    C++ 1 1

  6. Hayden727.github.io Hayden727.github.io Public

    CSS