Yun Chen
Researcher at Waabi
Old Toronto, Ontario, Canada
Summary
🤩
Rockstar🎓
Top SchoolYun Chen is a researcher and machine learning engineer with 11 years of experience, currently working at Waabi after research roles at Uber and internships at Alibaba and other tech organizations. He focuses on computer vision and motion forecasting, contributing to notable projects like a simplified Faster R-CNN implementation and the ECCV2020 LaneGCN repository where he improved training and evaluation pipelines. Yun maintains a hands-on open-source presence—his PyTorch tutorial/book repo modernizes legacy code and demonstrates practical deep learning projects from image captioning to style transfer. Based in Old Toronto and enrolled in a PhD program in Computer Science, he blends strong academic foundations with production-oriented research. An understated strength is his penchant for improving reproducibility and tooling across training loops, data preprocessing, and visualization.
11 years of coding experience
2 years of employment as a software developer
Bachelor's degree, Telecommunication Engineering, Bachelor's degree, Telecommunication Engineering at Beijing University of Posts and TelecommunicationsDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at 加拿大多伦多大学
English
Github Skills (19)
pytorch10
python10
machine-learning10
mask-rcnn10
deeplearning-ai10
deep-learning10
trainings10
object-detection10
faster-rcnn10
tensor10
graph-neural-network10
modeling10
artificial-intelligence9
data-preprocessing9
image-classification8
Programming languages (8)
Github contributions (5)
chenyuntc/simple-faster-rcnn-pytorch
Dec 2017 - Jul 2020
A simplified implemention of Faster R-CNN that replicate performance from origin paper
Role in this project:
ML Engineer
ML EngineerContributions:83 commits, 3 PRs, 73 pushes in 2 years 7 months
Contributions summary:Yun primarily contributed to the core training process of a Faster R-CNN model. Their commits reveal changes to the training loop, loss calculations, and model optimization, including modifications to the optimizer and learning rate. They also made adjustments to the data preprocessing steps and visualization tools.
pytorchdeep-learningr-cnnsimplifiedfaster
uber-research/LaneGCN
Jun 2020 - Apr 2021
[ECCV2020 Oral] Learning Lane Graph Representations for Motion Forecasting
Role in this project:ML Engineer
ML EngineerContributions:13 commits, 1 PR, 3 pushes in 10 months
Contributions summary:Yun primarily contributed to the training and evaluation pipeline for a motion forecasting model. They made changes to the training script (`train1.py`), likely for improved training procedures, and test script for submission. They updated the license, and added video/slides as part of the project. Additionally, they modified the data preprocessing steps to ensure proper data loading and processing.
pytorchforecastingeccv2020representationsmotion
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.