Hongyi Sun
Software Engineer DoorDash Labs
Sunnyvale, California, United States
Summary
🤩
Rockstar🎓
Top SchoolHongyi Sun is a software engineer with nine years’ experience specializing in autonomous driving prediction and planning, currently at DoorDash Labs after leading city-scale NGP/XNGP planning at XPENG. He has deep hands-on expertise in ML-driven trajectory and behavior prediction as well as ego-vehicle planning using sampling and optimization, developed during multi-year contributions to Baidu’s Apollo platform. Hongyi’s open-source work on Apollo includes MLP improvements, LSTM integration, and data-format tooling (.bin/.h5) that helped productionize models for real-world driving stacks. Trained at the University of Michigan with a dual-background B.E. in Electrical Engineering and Economics from Sichuan University, he blends rigorous engineering with system-level product thinking. Comfortable moving between research-quality ML models and pragmatic production engineering, he’s known for debugging initialization issues and bridging data pipelines to model deployment. Based in Sunnyvale, he brings a track record of shipping autonomous-driving features across diverse urban environments in China and the U.S.
9 years of coding experience
6 years of employment as a software developer
Bachelor of Engineering (B.E.) Double major in Electrical Engineering and Economics, Bachelor of Engineering (B.E.) Double major in Electrical Engineering and Economics at Sichuan UniversityMaster's degree Electrical and Computer Engineering, Master's degree Electrical and Computer Engineering at University of Michigan
English, Chinese
Github Skills (9)
mle10
pytorch10
deep-learning10
tensorflow10
ml10
autonomous-driving9
mlmodel9
data-pipeline8
data-pipelines8
Programming languages (5)
Github contributions (5)
ApolloAuto/apollo
Aug 2018 - Sep 2020
An open autonomous driving platform
Role in this project:
ML Engineer
ML EngineerContributions:5 reviews, 449 commits, 317 PRs in 2 years 1 month
Contributions summary:Hongyi primarily contributed to the Apollo autonomous driving platform by implementing and debugging machine learning models. Their work focused on improvements to an MLP (Multi-Layer Perceptron) model, including initialization bug fixes and the addition of an LSTM (Long Short-Term Memory) model. The user was also involved in converting data formats for model training, including generating and processing files in .bin and .h5 formats.
autonomousmachine-learningautonomyautonomous-drivingapollo
ManiiXu/apollo
May 2019 - May 2019
An open autonomous driving platform
Contributions:14 commits in 2 days
autonomousmachine-learningautonomous-drivingmavlinkdriving
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.