Brian Lee
Senior Software Engineer at Cascade Energy
Portland, Oregon, United States
Summary
🤩
Rockstar🎓
Top SchoolBrian Lee is a Senior Software Engineer based in Portland, Oregon with 14 years of experience building reliable systems across cloud infrastructure and applied machine learning. He spent seven years at AWS working on GameLift, EC2, and EBS, bringing production-grade backend and distributed-systems expertise to performance-sensitive services. Brian is an active open-source contributor with deep interest in reinforcement learning and search—his work on readable AlphaGo/AlphaGoZero implementations (including MuGo and contributions to TensorFlow Minigo) shows a knack for translating research ideas into testable, maintainable code. He also contributed to ML lifecycle tooling in MLflow and to TensorFlow documentation, indicating strength in both engineering and developer experience. Known for pragmatic refactors, improved logging, and MCTS enhancements, he combines hands-on coding with attention to observability and testability.
14 years of coding experience
9 years of employment as a software developer
Computer Science, Computer Science at Washington State University VancouverBachelor of Science (B.S.) in Computer Science Computer Science, Bachelor of Science (B.S.) in Computer Science Computer Science at Western Washington University
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (16)
algorithm10
algorithms10
machine-learning10
mlflow10
tensorflow10
game-development10
python10
ai10
protobuffer10
protobuf10
documentation10
implement9
deep-learning8
unit-testing8
testing8
Programming languages (8)
Github contributions (5)
brilee/MuGo
May 2016 - Mar 2020
Replicating AlphaGo's architecture in a readable manner
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:2 releases, 223 commits, 2 PRs in 3 years 10 months
Contributions summary:Brian contributed extensively to a project replicating AlphaGo's architecture. Their commits demonstrate the development of a tic-tac-toe game implementation and several strategies to play it. Key features include a framework for running game strategies, an interactive player, a random player, minimax and Monte Carlo tree search players. They were also involved in adding unit tests for the game logic.
mannerarchitecturereadablealphago
tensorflow/minigo
Dec 2017 - Mar 2020
An open-source implementation of the AlphaGoZero algorithm
Role in this project:Back-end Developer
Back-end DeveloperContributions:151 commits, 54 PRs, 25 pushes in 2 years 3 months
Contributions summary:Brian primarily focused on refactoring and improving the code related to the SGF (Smart Game Format) wrapper and MCTS (Monte Carlo Tree Search) node description within the AlphaGoZero algorithm implementation. They refactored the logging to take arbitrary comments, added more descriptive logging to each node in SGF, and implemented parent Q initialization, likely to improve the search performance. Furthermore, the user's commits also included adding functionality to handle the virtual losses within MCTS to enhance search quality.
pythondeep-learningmachine-learningmonte-carlo-tree-searchgomoku
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.