I am an AI Research Scientist at META, building the RL system for Muse Spark. Before that, I was a Postdoc Researcher in Sky Lab at UC Berkeley, worked with Prof. Ion Stoica. I received my Ph.D. in Computer Science from Tsinghua University in 2025, advised by Prof. Jidong Zhai in the PACMAN group. I received my Bachelor degree in Computer Science from Tsinghua University in 2021. I was a visiting student researcher at UC Berkeley in 2024.

Main Projects

I am building the large-scale RL system for Muse Spark, the latest LLM at META, and lead substantial efforts of its RL inference system.

I also work on vLLM, an open-source LLM inference engine.

I used to do research on deep learning compilers, quantum circuit simulation and general parallel computing. I was the lead of Tsinghua Student Cluster Competition team, and won several world championships.

Selected Publications

  • [SOSP’25] Jenga: Effective Memory Management for Serving LLM with Heterogeneity
    Chen Zhang, Kuntai Du, Shu Liu, Woosuk Kwon, Xiangxi Mo, Yufeng Wang, Xiaoxuan Liu, Kaichao You, Zhuohan Li, Mingsheng Long, Jidong Zhai, Joseph Gonzalez, Ion Stoica
    [Paper] [Code]
  • [ATC’24] MagPy: Compiling Eager Mode DNN Programs by Monitoring Execution States
    Chen Zhang, Rongchao Dong, Haojie Wang, Runxin Zhong, Jike Chen, Jidong Zhai
    [Paper] [Code]
  • [OSDI’23] Cocktailer: Analyzing and Optimizing Dynamic Control Flow in Deep Learning
    Chen Zhang, Lingxiao Ma, Jilong Xue, Yining Shi, Ziming Miao, Fan Yang, Jidong Zhai, Zhi Yang, Mao Yang
    [Paper] [Code]
  • [SC’22] UniQ: A Unified Programming Model for Efficient Quantum Circuit Simulation
    Chen Zhang, Haojie Wang, Zixuan Ma, Lei Xie, Zeyu Song, Jidong Zhai
    [Paper] [Code]
  • [ICS’21] HyQuas: Hybrid Partitioner Based Quantum Circuit Simulation System on GPU
    Chen Zhang, Zeyu Song, Haojie Wang, Kaiyuan Rong, Jidong Zhai
    Win the Best Student Paper award
    [Paper] [Code]
  • [TPDS’21] Critique of “Planetary Normal Mode Computation: Parallel Algorithms, Performance, and Reproducibility” by SCC Team From Tsinghua University
    Chen Zhang, Chenggang Zhao, Jiaao He, Shengqi Chen, Liyan Zheng, Kezhao Huang, Wentao Han, Jidong Zhai
    [Paper] [Code]

Honors

  • Special-Class Scholarship of Tsinghua University (Top 0.3%) 2020.11
    • The highest honor for undergraduates at Tsinghua, only 10 students per year
  • Beijing Outstanding Undergraduate Thesis (Top 0.5%) 2021.06
  • CCF Outstanding Undergraduate: 98 students in China 2020.10

Competitions

  • ISC21 Student Cluster Competition: Champion & Best Linpack (team captain) 2021.06
  • SC20 Student Cluster Competition: Champion & Best Linpack (team captain) 2020.11
  • SC19 Student Cluster Competition: Champion 2019.11
  • China Collegiate Programming Contest (Girls Only): Champion, solve all problems 2019.05
  • The 2018 ICPC Asia Jiaozuo Regional Contest: Rank 2, the first to solve problem C 2018.11
  • National Olympiad in Informatics: Silver medal 2016.07

Talks

Cocktailer: Analyzing and Optimizing Dynamic Control Flow in Deep Learning

  • OSDI2023, ChinaSys 2023Spring, HPCChina 2023, ETH AST Lab 2024

MagPy: Compiling Eager Mode DNN Programs by Monitoring Execution States

  • ATC2024, ChinaSys 2024Spring

HyQuas: hybrid partitioner based quantum circuit simulation system on GPU

  • ICS2021, HPCChina2022

UniQ: a unified programming model for efficient quantum circuit simulation

  • SC2022, HPCChina2022

Experience on Student Cluster Competitions

  • HPCChina 2021

Internship

  • Microsoft Research Asia, 2020.07-2021.12 (Mentors: Fan Yang and Lingxiao Ma)
  • Tencent Wechat Group, 2019.07-2019.08