Julius Frost

Machine Learning Engineer II at Cambium Assessment

Boston, Massachusetts, United States
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
🎓
Top School
Julius Frost is a Machine Learning Engineer II in Boston with eight years of engineering experience and dual CS degrees (BA/MS) from Boston University. He combines academic research in deep learning and reinforcement learning with production ML work at Cambium Assessment and MORSE Corp, shipping practical fixes and enhancements to RL tooling. An active contributor to the high-profile Ray project—focused on RLlib—he’s resolved device-mismatch bugs, improved offline dataset APIs, and hardened training/rollout scripts, showing an eye for robustness in distributed ML systems. Julius is particularly interested in the societal impact of AI and seeks roles that tackle hard, meaningful problems; colleagues describe him as pragmatic, research-fluent, and effective at bridging prototypes to production.
code8 years of coding experience
job5 years of employment as a software developer
bookBachelor of Arts - BA, Bachelor of Arts - BA at Boston University
languagesEnglish, Korean
github-logo-circle

Github Skills (11)

pytorch10
machine-learning10
ray10
rllib10
python10
reinforcement-learning10
deeplearning-ai9
sac9
deep-learning9
data-science7
tensorflow5

Programming languages (13)

C#JavaC++CRustGoHTMLShell

Github contributions (5)

github-logo-circle
ray-project/ray

Aug 2020 - Jul 2022

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.
Role in this project:
userML Engineer
Contributions:9 reviews, 14 commits, 15 PRs in 1 year 11 months
Contributions summary:Julius primarily contributed to the RLlib library, focusing on bug fixes and enhancements within the reinforcement learning domain. They addressed issues related to device mismatches in PyTorch, corrected tuple observation space handling in SAC, and improved the input API for custom offline datasets. The user also implemented fixes for D4RL paths and made improvements to the training and rollout scripts.
aimachine-learningraydistributedparallel
BU class registration bot that beats the competition
Contributions:26 commits, 23 pushes, 1 branch in 2 months
boston-universityregistrationbot
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.
Request Free Trial