Baruch Tabanpour
Staff Research Engineer at Google DeepMind
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolBaruch Tabanpour is a Staff Research Engineer based in Mountain View with 11 years of experience building machine learning and physics simulation systems. He progressed from ML roles at Squarespace and Google to senior research engineering at DeepMind, where he focuses on robotic simulation and tooling like MuJoCo, MJX and brax. His open-source contributions include improving MuJoCo integrations in JAX—fixing joint dynamics and adding visual and sensing features such as ray-mesh, sites, cameras and heightfields—helping bridge simulation fidelity and differentiable ML workflows. Trained in applied physics, applied math and electrical engineering, he brings a research-informed engineering approach that balances numerical rigor with production-grade software. Colleagues rely on him for complex simulation debugging and for shipping components that make advanced robotics research reproducible and scalable.
11 years of coding experience
8 years of employment as a software developer
Bachelor of Engineering (B.E.), Electrical Engineering, Physics, Bachelor of Engineering (B.E.), Electrical Engineering, Physics at The City College of New YorkMS, Applied Physics and Applied Math, MS, Applied Physics and Applied Math at Columbia UniversityMacaulay Honors College at The City University of New York
French
Github Skills (11)
simulations10
simulation10
mujoco10
jax10
physics10
python10
simulator10
machine-learning8
algorithm7
algorithms7
dir7
Programming languages (3)
Github contributions (5)
google-deepmind/mujoco
Oct 2022 - Jan 2026
Multi-Joint dynamics with Contact. A general purpose physics simulator.
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:4 releases, 5 reviews, 4 PRs in 3 years 3 months
Contributions summary:Baruch made several commits focused on improving the MuJoCo physics simulation within the context of JAX. They addressed bugs in the joint range, and also worked on including functionalities such as site, camera, and hfield, indicating expansion of the simulation capabilities. The user's work includes implementing a ray-mesh, suggesting contributions towards visual elements and simulation enhancements.
physics-enginephysicsroboticsrobotics-simulationbullet-physics
btaba/yarlp
Feb 2017 - Feb 2018
yet another reinforcement learning package
Contributions:91 commits, 2 PRs, 100 pushes in 1 year
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.