I am a Ph.D. student at UW, working with Carlos Guestrin. I am a member of SAMPL Lab and MODE Lab. I received my Master and Bachelor's degree from Shanghai Jiao Tong University, where I was a member of the ACM Class in SJTU.


I am interested in the intersection of machine learning and systems. The real excitement of this area comes from what can enable when we bring advanced learning techniques and system together. On that end, I am also pushing the direction on deep learning, knowledge transfer and lifelong learning.

Recently, I am working on TVM stack. Together with collaborators from SAMPL Lab, we co-design systems, machine learning algorithms, architecture, and programming language to enable future intelligent systems.

I build machine learning systems that are widely adopted. Besides TVM, the two other noticable ones are:

  • XGBoost, a scalable tree boosting system.
    It is one of the de-facto tools used by data scientists daily. You can find it in the production pipelines of the many major companies such as Uber, Airbnb, Amazon and Google Cloud. It has also been used to predict interesting physics events in LHC.
  • Apache MXNet(co-creator)
    One of the major deep learning frameworks, it is currently adopted by AWS.

I am graduating in 2019, and am on the job market this year.


  • University of Washington, 2013 - present
    Ph.D. student with Carlos Guestrin.

  • Google Brain. Summer 2015
    Research Intern with Ian Goodfellow.

  • Graphlab Inc. Summer 2014
    Intern, main author of boosted tree and neural net toolkit in Graphlab Create.

  • Apex Data and Knowledge Management Lab, Shanghai Jiao Tong University, China. 2010-2013
    Master Student with Yong Yu

  • System Biology Lab, Shanghai Jiao Tong University. 2010 - 2013
    Visiting Student, with Ping Ao

  • Huawei Noah's Ark Lab, Hong Kong. Summer 2012
    Research Intern, with Hang Li and Qiang Yang

  • Digital Enterprise Research Institute, Galway, Ireland. Winter 2011
    Visiting Scholar.

  • Microsoft Research Asia, Beijing, China. Summer 2009.
    Research Intern, with Jun Yan.


  • Google Ph.D. Fellowship
  • Champion in of KDD Cup 2012, track1, team leader.
  • 3rd place in KDDCup 2011, joint team leader.

Activity and Services

  • Organizer of the LearningSys workshop, NIPS 2015.