Scott Lee
Software Engineer at Together AI
United States
Summary
🤩
Rockstar🎓
Top SchoolScott Lee is a software engineer with 11 years of experience specializing in distributed data systems and ML infrastructure, currently working on LLM inference at Together AI and previously contributing as a Ray Data committer at Anyscale. He has deep hands-on expertise improving Ray's Dataset APIs—adding stats logging, schema unification for ragged Arrow arrays, and enhanced Parquet read filters—which reflects a knack for bridging data-science needs with robust backend implementations. Past roles at Lyft and Notion involved growth, marketplace, and product-focused engineering, showing he thrives across both research and product environments. A Berkeley MS/BS in EECS, he combines academic rigor with practical open-source impact on a widely used AI compute engine, and prefers collaborative problem-solving on complex systems.
11 years of coding experience
5 years of employment as a software developer
Master of Science - MS Electrical Engineering & Computer Science, Master of Science - MS Electrical Engineering & Computer Science at University of California, Berkeley
English, Spanish, Korean
Github Skills (8)
parquet10
data-pipeline10
data-pipelines10
pyarrow10
python10
data-science9
data-analysis9
machine-learning8
Programming languages (4)
Github contributions (5)
ray-project/ray
Nov 2022 - 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:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:796 reviews, 17 commits, 270 PRs in 2 months
Contributions summary:Scott focused on enhancing the Ray Data library, primarily within the `Dataset` and related APIs. Their contributions involved adding features such as automatically printing Dataset statistics and persisting them to log files. They also worked on schema unification for Datasets with ragged Arrow arrays and corrected issues with the implementation of core methods like `Dataset.count()`. Furthermore, the user added filter and column selection capabilities to the read_parquet and related APIs.
pythonconsistsruntimetensorflowserving
scottjlee/fidobot
Sep 2019 - May 2020
A friendly Slackbot for teaching staff support
Contributions:10 commits, 2 PRs, 8 pushes in 8 months
teachingslackbotslack-botslackstaff
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.