open_source_inference · hpc · ai_systems

High-performance AI infrastructure.Built in the open.

I build and optimize AI inference systems, using GitHub to share engineering experiments, open-source contributions and production-minded systems work.

View GitHub profile

02 · Profile signals

The engineer behind the repositories.

GitHub profile, public output and recent engineering activity—framed by a focus on high-performance AI systems.

Shijin Zhang

@Dovis01

Postgraduate student at Imperial College London (MSc Advanced Computing). Currently working at Z.AI.

Imperial College LondonLondonJoined Dec 1, 2020

Contributions / yr

Repos

Followers

Following

Pull Requests

last 90 days

Code Reviews

last 90 days

Issues & Comments

last 90 days

Active Projects

last 90 days

03 · Contribution activity

Open-source activity in motion.

A year of contributions, reviews, issues and active projects across the systems work receiving the most attention.

Contribution data unavailable

Recently active in

sgl-project/sglang vllm-project/vllm huggingface/transformers zai-org/GLM-5

04 · Repository index

Projects, experiments and systems in code.

Public repositories, languages and recent updates that show the work as it evolves on GitHub.

7 public repositories

sglangFork

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

Python1mo ago

transformersFork

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

Python2mo ago

FlashMLAFork

FlashMLA: Efficient Multi-head Latent Attention Kernels

3mo ago

vllmFork

A high-throughput and memory-efficient inference and serving engine for LLMs

3mo ago

hermes-agent

The agent that grows with you

3mo ago

vllm-omniFork

A framework for efficient model inference with omni-modality models