Qiangeng Xu
Research Scientist at Waymo
Mountain View, California, United States
Summary
👤
Senior🎓
Top SchoolQiangeng (Charlie) Xu is a research scientist with 11 years of experience at the intersection of 3D vision, generative models, and neural rendering, currently at Waymo after research roles at Google AR. He holds a PhD in Computer Science from USC and has contributed to influential projects and papers such as Point-NeRF, bringing practical ML engineering skills like CUDA kernel development and COLMAP point-cloud integration to state-of-the-art neural radiance field systems. His background spans aerospace engineering to large-scale industry systems—experience that informs his work on digital humans, 3D reconstruction, and novel learning schemes like contrastive and meta-learning. Notably, he has moved research prototypes toward production-readiness, optimizing performance-critical components and adapting models across diverse datasets.
11 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at University of Southern CaliforniaBachelor of Engineering (B.E.) Aerospace Aeronautical and Astronautical Engineering, Bachelor of Engineering (B.E.) Aerospace Aeronautical and Astronautical Engineering at Beihang UniversityHigh School attached with TsinghuaMaster of Science (M.S.), Master of Science (M.S.) at University of Pennsylvania
Chinese, English
Github Skills (11)
cuda10
pytorch10
machine-learning10
point-cloud10
3d-reconstruction10
python9
render9
rendering9
computer-vision9
c-language8
cprogramming-language8
Programming languages (4)
Github contributions (5)
Xharlie/pointnerf
Feb 2022 - Nov 2022
Point-NeRF: Point-based Neural Radiance Fields
Role in this project:
ML Engineer
ML EngineerContributions:41 commits, 1 PR, 29 pushes in 9 months
Contributions summary:Qiangeng contributed to the Point-NeRF project by adding functionality related to COLMAP point cloud input, enhancing the model's ability to incorporate external point data. They also made modifications to scripts, including testing and training configurations, specifically adapting them for different datasets and model parameters. Furthermore, the user implemented CUDA kernels for efficiently querying and processing point cloud data, including calculations for ray position masking and neighbor identification. These updates indicate a focus on improving the model's performance and integration with external data sources.
scene-reconstructionradiancepoint-cloudvolume-renderingnerf
Xharlie/Grid-GCN
Mar 2020 - Mar 2020
Contributions:35 commits, 6 pushes, 1 branch in 17 days
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.