Stephen Koo
Staff Software Engineer at Google
California, United States
Summary
👤
Senior🎓
Top SchoolStephen Koo is a Staff Software Engineer with 13 years of experience building AI and infrastructure at scale, currently leading core platform work for Google Workspace after progressing through Gmail, Chat, and Speech teams. He combines deep ML research experience—developing sequence-to-sequence models for long-form conversational speech—with hands-on full-stack engineering, having rebuilt CodaLab’s backend and shipped UI/backend features like an enhanced web terminal autocomplete. Comfortable across microservices, REST APIs, and AI infrastructure, he has a track record of stabilizing projects (120+ GitHub issues closed on CodaLab) and shipping zero-downtime migrations. Based in California and trained at Stanford in AI, he blends research rigor with pragmatic production engineering and an eye for developer experience.
13 years of coding experience
9 years of employment as a software developer
Master of Science (M.S.) Computer Science (Artificial Intelligence), Master of Science (M.S.) Computer Science (Artificial Intelligence) at Stanford University
English, Chinese, Japanese, German, Hindi
Github Skills (13)
command-line-interface10
autocomplete10
javascript10
front-end-development10
react9
backend9
back-end-development9
tournament9
api8
web-framework8
apidoc8
python8
machine-learning5
Programming languages (4)
Github contributions (5)
codalab/codalab-competitions
Oct 2015 - Dec 2015
CodaLab Competitions
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:63 commits, 26 PRs, 32 pushes in 1 month
Contributions summary:Stephen primarily worked on improving the web terminal functionality within the CodaLab Competitions repository. Their contributions included implementing autocomplete features using JavaScript, enhancing command execution logic, and modifying the user interface to respond to backend actions. Furthermore, the user refactored existing autocomplete code and addressed issues related to command processing and display. The user also made modifications to the Python backend related to the autocomplete functionality.
benchmarksartificial-intelligencesciencemachine-learning
kashizui/codalab-cli
Sep 2015 - May 2016
Contributions:203 pushes, 105 branches in 7 months
bundlecommand-line-interfaceworksheetssupportingcli
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.