Richard Soong
Data Engineer
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolRichard Soong is a data engineer and UC Berkeley Data Science undergraduate with a human-centered approach to applying analytics at the intersection of technology and society. He combines academic experience—teaching Data 8 labs and mentoring literacy programs—with hands-on work in industry and research, including NLP safety models and economic analyses of cryptocurrency. Richard has contributed to high-profile open-source distributed ML projects, integrating Horovod with Ray for scalable training and adding HyperOpt support to Ray Tune, demonstrating practical MLOps and backend engineering chops. His roles span data science, product-focused analytics, and full-stack reinforcement learning tutorials, reflecting both breadth and execution. Based in the Bay Area, he pairs a curiosity for sports statistics and food with a track record of shipping reproducible ML infrastructure.
11 years of coding experience
3 years of employment as a software developer
Bachelor's degree, Data Science, Bachelor's degree, Data Science at University of California, BerkeleyLowell High SchoolMaster of Science - MS, Analytics: Data Science Emphasis, Master of Science - MS, Analytics: Data Science Emphasis at Georgia Institute of Technology
Chinese, English
Github Skills (19)
javascript10
distributed-training10
python10
machine-learning10
horovod10
reinforcement-learning10
distributed-computing10
hyperparameter-optimization10
ray10
apim9
api9
deep-learning9
game-development9
tensorflow8
pytorch8
Programming languages (11)
Github contributions (5)
ray-project/tutorial
Oct 2018 - Jul 2020
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:34 commits, 50 PRs, 43 pushes in 1 year 9 months
Contributions summary:Richard primarily worked on updating and improving the rllib exercises, specifically focusing on the JavaScript-based Pong game. Their contributions involved modifying the game's JavaScript code, likely to implement or refine features related to reinforcement learning exercises and interactions with an external API. The changes suggest a focus on integrating the game with a reinforcement learning framework or environment.
ray-project/ray
Jan 2017 - Jan 2023
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:ML Engineer & Backend Developer
ML Engineer & Backend DeveloperContributions:2 releases, 3188 reviews, 604 commits in 6 years
Contributions summary:Richard contributed to the `ray-project/ray` repository by implementing HyperOpt support (v2) within the Tune module, modifying Python files within the "ray/tune" directory. Their code focused on integrating HyperOpt, a library for hyperparameter optimization, into Ray's distributed computing engine. The changes include modifications to the "python/ray/tune/hpo_scheduler.py" and "python/ray/tune/trial_runner.py" files. Additionally, the user polished docs in the "doc/source/tune.rst" files.
pythonconsistsruntimetensorflowserving
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.