John Balis
Platform Engineer
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolJohn Balis is a platform engineer with nine years of software engineering experience, currently based in Mountain View and focused on building reliable infrastructure and tooling. He combines academic research in reinforcement learning from UW–Madison with practical platform work at Matic Robots, bridging experimental ML systems and production-ready engineering. An active open-source contributor, he improved core Gym functionality—adding return_info support and auto-reset behavior—to make RL environments more robust for developers and researchers. He has hands-on experience across data access tools, robotics planning, and scientific software from internships and research roles, reflecting a strong generalist foundation. Known for pragmatic problem-solving, he seeks roles that blend platform engineering with open-source AI work.
9 years of coding experience
2 years of employment as a software developer
Master's degree (Computer Sciences), Computer Science, Master's degree (Computer Sciences), Computer Science at University of Wisconsin-Madison
English, Japanese
Github Skills (9)
gymnasium10
openai-gym10
environmental10
dev-environment10
environ10
python10
reinforcement-learning10
enviroment10
unit-testing9
Programming languages (12)
Github contributions (5)
openai/gym
Feb 2022 - Aug 2022
A toolkit for developing and comparing reinforcement learning algorithms.
Role in this project:
ML Engineer
ML EngineerContributions:47 reviews, 6 commits, 8 PRs in 6 months
Contributions summary:John primarily contributed to the Gym library by implementing and testing features related to the `reset` function, adding the ability to return an info dictionary. They modified the `async_vector_env.py`, `sync_vector_env.py`, and wrapper files to integrate the new `return_info` functionality. Additionally, the user implemented an auto-reset wrapper and made changes related to the handling of environment seeding, as well as removing the deprecated `seed` function from the vector environments.
comparingreinforcement-learning-algorithmsdevelopingdeep-learningreinforcement-learning
balisujohn/mltests
Sep 2018 - Apr 2020
This is a project with the loose goal of finding interesting ways to generate intelligent agents through trial and error.
Contributions:131 commits, 10 PRs, 113 pushes in 1 year 7 months
intelligent-agentsgoalinterestingfindingtrial-and-error
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.