Yu Zhu
Software Engineer at Waymo
California, United States
Summary
🤩
Rockstar🎓
Top SchoolYu Zhu is a software engineer with 13 years of experience specializing in perception systems for autonomous vehicles, currently building perception at Waymo after prior roles at XPENG, Aptiv, and DENSO. He combines applied research instincts from an MS in Robotics with hands-on production engineering, shipping algorithms that bridge sensors to real-time systems. An active contributor to ML tooling, he has implemented Swin Transformer components in the PaddleViT project, demonstrating deep familiarity with visual transformers and low-level model internals. Based in California, he enjoys implementing AI algorithms from scratch and often moves ideas from prototype code to deployable systems, making him adept at both experimentation and robust engineering.
13 years of coding experience
2 years of employment as a software developer
Master of Science - MS, Robotics, Master of Science - MS, Robotics at University of MichiganBachelor of Engineering - BE, Bachelor of Engineering - BE at Dalian University of Technology
Github Skills (8)
swin-transformer10
paddlepaddle10
computer-vision10
transformer10
deep-learning10
python10
image-classification10
pytorch3
Programming languages (3)
Github contributions (5)
BR-IDL/PaddleViT
Jun 2021 - Sep 2022
:robot: PaddleViT: State-of-the-art Visual Transformer and MLP Models for PaddlePaddle 2.0+
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:2 releases, 35 reviews, 416 commits in 1 year 2 months
Contributions summary:Yu's contributions center on implementing Swin Transformer models within the PaddlePaddle framework, indicating a focus on computer vision and deep learning. The commits involve the addition and modification of code related to Swin Transformer, including core components such as PatchEmbedding, Attention layers, and the Encoder. The user's work appears to be directly related to the implementation of state-of-the-art visual transformer models for image classification, demonstrating involvement in both back-end model development and machine learning engineering.
mlpsemantic-segmentationstate-of-the-artvisual-artclassification
xperzy/facelabelpython
Sep 2014 - Oct 2014
Contributions:9 commits in 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.