My role at DeepSeek-AI is dedicated to help building the training and serving infrastructure for the next generation of DeepSeek models. A key part of my work is optimizing our large-scale internal software systems to ensure peak performance on various hardware platforms.

I work with the most excellent peers and engineers to build incredible inference system for SOTA large language models like DeepSeek-R1. Our system deployed for WeChat serves a billion of users.

We also work closely with open-source communities, such as SGLang and vLLM, to contribute our industrial insight and optimizations for extreme large deployment.

I worked on the Application Defined Network project under the supervision of Xiangfeng and Ratul. I independently developed a compiler backend that compiles our AppNet program into Envoy Native Filter.

See also: FOCI Center, UW Syslab.

I am fortunate to work alongside many senior SDEs. We worked hard to build a high-performance RDMA library and a distributed in-memory cache for machine learning, which are deployed on hundreds of high-end GPU machines with eight GPUs each. I learned a lot from the industrial experience.

I worked closely with Jianyu under the guidance of Ting Cao. I contributed some preliminary results to ISCA'2024 Pre-gated MoE: An Algorithm-System Co-Design for Fast and Scalable Mixture-of-Expert Inference. But unfortunately, my work is completed at a relatively early stage of this project, which was not sufficient to earn authorship.

During high school, I focused extensively on Olympiads in Informatics (OI), which brings me a very good understanding in algorithms and data structures. I won a silver medal in Chinese National Olympics in Informatics Competition (NOI 2020), and several ICPC regions and semi-final gold medals during my bachelor time.