Saran Tunyasuvunakool
Senior Staff Research Engineer at Google DeepMind
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolSaran Tunyasuvunakool is a Senior Staff Research Engineer at Google DeepMind who leads the Robotics Simulation Team and oversees MuJoCo development and much of DeepMind's open sourcing efforts. With 11 years of experience bridging research and production, he combines deep expertise in high-performance C++ and SIMD-optimised scientific code from a PhD in numerical general relativity with day-to-day stewardship of a widely used physics simulator. He regularly ships backend improvements, performance optimisations, and robust native resource handling across key repos like mujoco and dm_control, and often surfaces value through careful documentation and tooling fixes. Comfortable on supercomputers and desktop platforms alike, he has ported legacy Fortran into vector-friendly C++ and tackled platform-specific quirks such as macOS VSync workarounds. His background simulating exotic black hole systems gives him a rare mix of theoretical physics intuition and pragmatic engineering that benefits robot learning and embodied AI. Based in London, he excels at turning complex numerical methods into reliable, production-grade simulation infrastructure.
11 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy (PhD) Numerical General Relativity, Doctor of Philosophy (PhD) Numerical General Relativity at University of CambridgeCertificate in Science Mathematics, Certificate in Science Mathematics at University of CanterburyHigh School, High School at Burnside High School
English, Thai
Stackoverflow
1,089reputation
62kreached
13answers
13questions
wifi
top-5%
Github Skills (24)
simulation10
c-language10
python10
simulator10
simulations10
plugin-development10
physics10
performance-optimization10
mujoco10
cprogramming-language10
documentation10
reinforcement-learning9
wifi9
refactoring8
refactor8
Programming languages (10)
Github contributions (5)
google-deepmind/mujoco
Oct 2021 - Jan 2023
Multi-Joint dynamics with Contact. A general purpose physics simulator.
Role in this project:
Back-end Developer & Documentation Specialist
Back-end Developer & Documentation SpecialistContributions:12 releases, 116 reviews, 112 commits in 1 year 3 months
Contributions summary:Saran primarily focused on addressing documentation issues and implementing bug fixes within the MuJoCo physics simulator. Their contributions included correcting license links, updating changelogs, and resolving code formatting issues. Furthermore, the user was involved in refactoring the low-level implementation details.
physics-enginephysicsroboticsrobotics-simulationbullet-physics
google-deepmind/dm_control
Jan 2018 - Dec 2022
Google DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo.
Role in this project:Back-end Developer
Back-end DeveloperContributions:16 releases, 7 reviews, 110 commits in 4 years 11 months
Contributions summary:Saran made significant contributions to the `dm_control` library, focusing on improvements related to MuJoCo physics simulation and reinforcement learning environments. Their work included adding methods for freeing native resources, implementing error handling, and enhancing the functionality of rendering contexts. The user also implemented a plugin mechanism for actuators and sensors, demonstrating a deep understanding of the underlying MuJoCo integration within the library. Furthermore, the user focused on performance optimizations within the simulation.
simulationreinforcementsoftware-stackphysics-based-simulationsimulation-based
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.