Adrien Ecoffet
Research Scientist at OpenAI
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolAdrien Ecoffet is a research scientist in deep reinforcement learning with 14 years of engineering and research experience, currently at OpenAI in San Francisco. He has a track record bridging applied ML and production systems from roles at Uber (Research Scientist, AI Resident) and Quora (Staff/SWE), and holds an MS in Computer Science from Georgia Tech. Adrien contributes to open-source RL tooling—most notably back-end work on the MineRL project enhancing multi-agent support and realistic agent dynamics—which reflects a focus on environment design and sample-efficient RL. He combines hands-on systems engineering (server-side and Java components) with algorithmic research, enabling scalable multi-agent experiments. Colleagues describe him as someone who translates complex research requirements into robust platform features that accelerate experimentation.
14 years of coding experience
5 years of employment as a software developer
EpitechUniversity of California, San DiegoMaster of Science - MS, Computer Science, Master of Science - MS, Computer Science at Georgia Institute of Technology
English, French
Github Skills (12)
javas10
environmental10
minecraft-forge10
dev-environment10
minecraft-fabric10
game-development10
environ10
reinforcement-learning10
java10
enviroment10
python9
api-design8
Programming languages (6)
Github contributions (5)
minerllabs/minerl
Sep 2020 - Jul 2021
MineRL Competition for Sample Efficient Reinforcement Learning - Python Package
Role in this project:
Back-end Developer
Back-end DeveloperContributions:32 reviews, 53 commits, 17 PRs in 10 months
Contributions summary:Adrien primarily contributed to the back-end aspects of the project, focusing on integrating multi-agent support and specifying health and food attributes. They modified core components, including environment specifications, multi-agent environment logic, and server-side Malmo components, likely written in Java. Additionally, the user implemented features related to agent health, food, and breaking speed, enhancing the game's dynamics. These changes suggest a focus on game environment and agent behavior within a reinforcement learning context.
pythonreinforcement-learning
uber-research/go-explore
Jul 2020 - Jan 2022
Code for Go-Explore: a New Approach for Hard-Exploration Problems
Contributions:1 release, 2 commits, 7 pushes in 1 year 6 months
go
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.