Matt B
Staff Software Engineer at DeepMind
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolMatt B is a Staff Software Engineer based in Mountain View with eight years of focused engineering experience and a broader career stretching back through leadership roles in technical architecture. He specializes in high-performance backend systems and physics simulation—contributing substantial refactors and parallelization work to prominent open-source engines like DeepMind’s MuJoCo and Bullet Physics. At DeepMind and previously X and Google he has driven system architecture and performance optimizations that bridge research-grade simulation and production-scale infrastructure. Trained in aerospace and mechanical engineering with an MS in computer science, he blends rigorous modeling intuition with practical systems design. Notably, his contributions to memory allocation, custom container types, and threading primitives enabled safer, faster simulation pipelines and asset handling in widely used physics projects.
7 years of coding experience
15 years of employment as a software developer
MS, Computer Science, MS, Computer Science at The George Washington UniversityBS, Aerospace Engineering, Mechanical Engineering, BS, Aerospace Engineering, Mechanical Engineering at Rensselaer Polytechnic Institute
Github Skills (22)
simulation10
robotics10
c-language10
simulator10
multithreading10
simulations10
memory-management10
c1110
c1710
physics10
thread10
hyper-threading10
urdf10
threaded10
game-development10
Programming languages (2)
Github contributions (5)
bulletphysics/bullet3
Aug 2018 - Aug 2022
Bullet Physics SDK: real-time collision detection and multi-physics simulation for VR, games, visual effects, robotics, machine learning etc.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:20 commits, 8 PRs, 2 comments in 4 years
Contributions summary:Matt contributed to the Bullet Physics SDK by implementing features related to Inverse Kinematics (IK) calculations, specifically allowing the user to specify current joint positions. They updated the URDF importer to process sensor elements, adding them as fixed-joint-attached elements and modifying header files for sensor support. Furthermore, the user integrated a simulation timestamp parameter, improving physics engine functionality. They made various code format changes to improve maintainability.
pybulletbullet-physicssimulationsimulatorreal-time
google-deepmind/mujoco
Mar 2024 - Apr 2024
Multi-Joint dynamics with Contact. A general purpose physics simulator.
Role in this project:Back-end Developer & System Architect
Back-end Developer & System ArchitectContributions:2 reviews in 1 month
Contributions summary:Matt's primary contribution focused on refactoring and optimizing memory allocation within the MuJoCo physics engine. They implemented a clean byte allocation interface and stack management for the `mjData` structure. The user also added new container types like `ArrayList` and threading primitives, and integrated them into the solver for the parallelization of some operations. Furthermore, the user made modifications in order to save normalizing transforms for mesh assets and added storing paths for hfields, textures and skins.
physics-enginephysicsroboticsrobotics-simulationbullet-physics
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.