Zun Li
Research Engineer at Google DeepMind
New York, New York, United States
Summary
👤
Senior🎓
Top SchoolZun Li is a research engineer at Google DeepMind with eight years of experience at the intersection of multi-agent systems, deep learning, and AI research. He completed a PhD at the University of Michigan and has a track record of turning cutting-edge research into practical algorithms and productionized systems—from extending AlphaZero for general-sum imperfect-information games to building C++ data pipelines at Google that handled 100GB-scale advertiser histories. His open-source contributions to DeepMind’s OpenSpiel include novel implementations of projected replicator dynamics, multiagent Q-learning, and Nash/Stackelberg solvers, reflecting deep expertise in game-theoretic RL. Comfortable bridging theory and engineering, he has published and evaluated systems against humans in negotiation domains and brings both academic rigor and production-grade delivery to large-scale AI problems.
9 years of coding experience
1 year of employment as a software developer
Doctor of Philosophy - PhD Computer Science and Engineering, Doctor of Philosophy - PhD Computer Science and Engineering at University of MichiganBachelor of Science - BS Computer Science, Bachelor of Science - BS Computer Science at Shanghai Jiao Tong University
Chinese, Chinese
Stackoverflow
1reputation
706reached
0answers
1question
Github Skills (12)
algorithms10
python10
reinforcement-learning10
videogames9
machine-learning9
coregames9
matrix-multiplication6
matrix6
linear-algebra6
numpy6
scientific-computing6
cpp5
Programming languages (4)
Github contributions (5)
google-deepmind/open_spiel
Jul 2020 - Jan 2023
OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.
Role in this project:
ML Engineer
ML EngineerContributions:17 reviews, 70 commits, 30 PRs in 2 years 6 months
Contributions summary:Zun contributed to the implementation of projected replicator dynamics algorithms within the open_spiel repository, specifically focusing on L2-projection operators and vectorized simplex projections. They also introduced and developed multi-agent reinforcement learning algorithms, including a tabular multiagent Q-learning agent, and included a matrix Nash solver. Furthermore, the user added implementations of Nash Averaging and Stackelberg equilibrium solvers.
reinforcement-learninggamesmultiagentcpppython
rezunli96/open_spiel
Jul 2020 - Oct 2023
OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.
Contributions:81 pushes, 15 branches in 3 years 3 months
reinforcement-learning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.