Shih-yang Su
Research Scientist at Waymo
United States
Summary
🤩
Rockstar🎓
Top SchoolShih-yang Su is a research scientist specializing in 3D computer vision and human digitalization, currently driving research at Waymo after multiple research internships at Meta and industry stints. With 11 years of experience spanning ML, RL, CV and MIR, he blends deep academic training (PhD candidate at UBC, MEng from Virginia Tech) with hands-on engineering—evidenced by contributions to the CVPR 2020 3D photo inpainting codebase where he integrated monocular depth estimation pipelines. He has applied probabilistic graph generation and GCNs in banking during a Borealis AI internship and has taught advanced ML and deep learning courses, demonstrating both research depth and pedagogy. Comfortable moving between prototype code and production-focused research, he often bridges novel models and practical system integration. Colleagues would note his cross-domain curiosity—from music information retrieval to autonomous driving—and a knack for adapting academic methods into applied engineering.
11 years of coding experience
1 year of employment as a software developer
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at The University of British ColumbiaMaster's degree Electrical and Computer Engineering, Master's degree Electrical and Computer Engineering at Virginia Tech College of Engineering學士 資訊工程學系, 學士 資訊工程學系 at National Tsing Hua University
Github Skills (6)
pytorch10
computer-vision10
python10
machine-learning8
image-processing8
bash7
Programming languages (1)
Github contributions (5)
vt-vl-lab/3d-photo-inpainting
Apr 2020 - Jun 2021
[CVPR 2020] 3D Photography using Context-aware Layered Depth Inpainting
Role in this project:
ML Engineer
ML EngineerContributions:7 commits, 2 PRs, 16 pushes in 1 year 2 months
Contributions summary:Shih-yang contributed to the integration of monocular depth estimation techniques within the 3D photo inpainting project. Their work included adding support for "BoostingMonocularDepth," integrating its outputs, and updating the download script to include necessary weights. They also modified the main execution script to incorporate the new depth estimation method and adjusted cropping functionality within image processing.
context-awarestructure-from-motioncvpr-2020inpaintingcomputer-vision
LemonATsu/LemonATsu.github.io
Sep 2016 - Sep 2022
Contributions:296 commits, 326 pushes, 1 branch in 6 years 1 month
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.