Kevin Lee
SWE at Google
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolKevin Lee is a software engineer at Google with eight years of experience building production AI and computer vision systems, currently contributing to Cloud AI and Vertex AI pipelines. He holds an MS in ECE from UCLA and has a track record of applied research—from lidar inpainting with GANs at MIT Lincoln Laboratory to action recognition work accepted at a CVPR workshop. Kevin blends research and engineering, shipping features in Kubeflow Pipelines (including ParallelFor over Artifact lists and dynamic machine type support) to make ML workflows more robust and flexible. His background spans embedded hardware automation at JPL to deep learning at HRL, giving him a strong systems-oriented perspective on deploying ML at scale. Based in Mountain View, he pairs hands-on implementation skills with a knack for turning research insights into production-ready tooling.
8 years of coding experience
2 years of employment as a software developer
Bachelor's degree, Electrical Engineering, Bachelor's degree, Electrical Engineering at University of California, Los Angeles
Github Skills (14)
kubeflow10
kubernetes10
machine-learning10
mlops10
python10
data-science10
pipeline10
kubernetes-pods10
tensorflow9
gcp8
cicd7
dockers7
testing7
docker7
Programming languages (2)
Github contributions (5)
kubeflow/pipelines
Sep 2023 - Jan 2025
Machine Learning Pipelines for Kubeflow
Role in this project:
ML Engineer & DevOps Engineer
ML Engineer & DevOps EngineerContributions:61 reviews, 18 PRs, 30 pushes in 1 year 3 months
Contributions summary:Kevin primarily focused on extending the Kubeflow Pipelines SDK by adding support for `dsl.ParallelFor` over lists of Artifacts. This involved significant code changes to the `for_loop` module, as well as the `pipeline_spec_builder` and compiler test. The user also worked on preventing the compilation of pipelines with incorrect usage of ParallelFor operations and added a test case for this scenario. Furthermore, the user contributed to supporting dynamic machine type parameters within custom training jobs.
pipelinetektondata-sciencemachine-learningmlops
KevinGrantLee/python-time-tracking
Feb 2021 - May 2021
Contributions:39 pushes, 1 branch, 1 comment in 3 months
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.