Jie Tan
Director at Georgia Institute of Technology
Mountain View, California, United States
Summary
👤
Senior🎓
Top SchoolJie Tan is a director and research leader at Google DeepMind with 14 years of experience building AI and robot systems that bridge foundation models, reinforcement learning, and physics-based simulation. He leads teams focused on robot mobility and embodied reasoning, translating cutting-edge research into sim-to-real robotic capabilities. His PhD from Georgia Tech produced four SIGGRAPH papers on physically-based character animation, and his early work at Lytro and contributions to DART and Bullet indicate deep expertise in simulation, contact solvers, and robust RL environments. Jie combines hands-on systems engineering (fixing core solvers and integrating RL stacks) with academic mentorship as an adjunct at Georgia Tech. He is known for tackling long-standing problems in graphics and robotics alike, bringing principled physics and learning methods to real-world robots. Based in Mountain View, he blends open-source impact with lab-scale innovation to push embodied AI toward practical deployment.
14 years of coding experience
11 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Computer Science, Doctor of Philosophy (Ph.D.) Computer Science at Georgia Institute of TechnologyMaster's Degree Computer Science, Master's Degree Computer Science at Shanghai Jiao Tong University
English, Chinese
Github Skills (19)
algorithm10
simulation10
algorithms10
c-language10
robotics10
python10
simulations10
reinforcement-learning10
data-structure10
dynamics10
tensorflow10
data-structures10
pybullet10
cprogramming-language10
kinematics9
Programming languages (4)
Github contributions (5)
bulletphysics/bullet3
May 2017 - Apr 2018
Bullet Physics SDK: real-time collision detection and multi-physics simulation for VR, games, visual effects, robotics, machine learning etc.
Role in this project:
ML Engineer
ML EngineerContributions:12 commits, 21 PRs in 11 months
Contributions summary:Jie contributed to the development of a reinforcement learning environment within the Bullet Physics SDK, specifically for the Minitaur robot. They focused on integrating Sonnet, a neural network library, into the agent implementation. The user also removed some potentially confusing data, and made changes to the PPO algorithm for better functionality. They also added a new URDF file for the minitaur quadruped robot.
pybulletbullet-physicssimulationsimulatorreal-time
dartsim/dart
Nov 2011 - Apr 2012
DART: Dynamic Animation and Robotics Toolkit
Role in this project:Back-end Developer
Back-end DeveloperContributions:7 commits in 4 months
Contributions summary:Jie primarily worked on the `Lemke` solver, a core component of the DART (Dynamic Animation and Robotics Toolkit). Their contributions involved fixing bugs related to the solver's crashing issues, adding debugging logs for cross-platform debugging, and fixing a compiling issue. Furthermore, they integrated an ODE LCP solver and implemented scaling and robustness improvements. These changes indicate a focus on improving the stability and reliability of a key computational element within the robotics and simulation domain.
kinematicsroboticsanimationcomputer-animationsimulation
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.