Lee Clement
Machine Learning Engineer at Roboflow
St. John's, Newfoundland and Labrador, Canada
Summary
🤩
Rockstar🎓
Top SchoolLee Clement is a Machine Learning Engineer with 11 years of experience blending robotics research and production ML, currently building Vision AI at scale at Roboflow. He previously led prediction and planning ML teams for L4 autonomous vehicles at Oxa, shipping multi-object tracking and motion prediction systems as a Staff Engineer and tech lead. Lee holds a Ph.D. in Mobile Robotics from the University of Toronto and has translated academic work on long-term visual navigation into practical AV stacks and course material for MOOCs. An active open-source contributor, he enhanced the widely used pykitti toolkit to improve KITTI image handling and odometry support, reflecting a pragmatic focus on data and tooling. Based in St. John's, Newfoundland, he combines deep research rigor with hands-on system architecture and people leadership across industry and academia.
11 years of coding experience
8 years of employment as a software developer
Bachelor of Science (B.Sc.) Physics Computer Science, Bachelor of Science (B.Sc.) Physics Computer Science at University of ManitobaDoctor of Philosophy (Ph.D.) Mobile Robotics, Doctor of Philosophy (Ph.D.) Mobile Robotics at University of Toronto
English, French
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (7)
computer-vision10
opencv10
python10
image-processing10
numpy10
robotics9
pandas8
Programming languages (5)
Github contributions (5)
utiasSTARS/pykitti
Apr 2016 - Feb 2019
Python tools for working with KITTI data.
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:40 commits, 7 PRs, 38 pushes in 2 years 11 months
Contributions summary:Lee contributed to the `pykitti` repository by adding and modifying code related to image loading and processing for the KITTI dataset. They implemented OpenCV integration to convert images to uint8 format for easier use, as well as refactored the code to use namedtuples and generators for enhanced efficiency. Furthermore, they added support for odometry benchmark data, extending the capabilities of the library to handle various KITTI dataset formats.
kittipythonkitti-datasetcomputer-visionrobotics
utiasSTARS/cat-net
Dec 2017 - Sep 2019
Canonical Appearance Transformations
Contributions:24 commits, 26 pushes, 2 branches in 1 year 10 months
cat-netdeep-learningcomputer-visionrobotics
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.