Matt Hoffman
Senior Staff Research Scientist at Google DeepMind
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolMatt Hoffman is a Senior Staff Research Scientist at DeepMind with 11 years of experience building research-oriented systems for reinforcement learning and large-scale LLM training using online and offline RL. His work emphasizes general, widely applicable RL algorithms and led to the creation and maintenance of Acme, a widely used open-source RL framework where he has driven documentation, tutorials, packaging, and build/test improvements. With a PhD from UBC and postdoctoral work at Cambridge, he brings deep expertise in continuous-action RL and Bayesian optimization for hyperparameter tuning. Based in London, he blends rigorous academic research with pragmatic software engineering to make advanced RL methods accessible and reproducible. An underappreciated strength is his focus on developer experience—ensuring research code is well-documented, versioned, and easy to adopt by the community.
11 years of coding experience
3 years of employment as a software developer
Bachelor of Science - BS, Computer Science and Mathematics, Bachelor of Science - BS, Computer Science and Mathematics at University of WashingtonDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at The University of British Columbia
Github Skills (7)
python10
reinforcement-learning10
setuptools9
documentation9
py9
agent9
testing8
Programming languages (2)
Github contributions (5)
google-deepmind/acme
May 2020 - Sep 2022
A library of reinforcement learning components and agents
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:3 releases, 4 reviews, 68 commits in 2 years 4 months
Contributions summary:Matt primarily contributed to the Acme library's documentation and setup. They updated the `setup.py` file to include more documentation, added a tutorial notebook example, and corrected the description for PyPI. The user also addressed versioning issues by bumping the version numbers and making explicit the versioning scheme. They also modified the test scripts and imports to improve the build and test process.
reinforcement-learningreinforcementagentsdeep-reinforcement-learning
mwhoffman/config
Jan 2015 - Feb 2025
Contributions:222 pushes in 10 years 2 months
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.