Sheng Zhong
Robotics Research Engineer at Sunday
San Jose, California, United States
Summary
🤩
Rockstar🎓
Top SchoolSheng Zhong is a Robotics PhD candidate at the University of Michigan with 11 years of software engineering experience, blending research-grade control theory with practical back-end systems work. He contributes to open-source projects ranging from FPGA CAD tooling (improving server-side logging and benchmarking in the widely used VTR flow) to fast, differentiable MPC solvers in PyTorch, where he builds example integrations using learned dynamics and gym environments. Based in Ann Arbor, he bridges robotics research and production engineering—comfortable instrumenting databases and servers for benchmarking as well as prototyping neural-network-driven controllers. Sheng’s profile reflects a pragmatic engineer who translates theoretical control ideas into reproducible code and reliable infrastructure for experiments.
11 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy - PhD, Robotics, Doctor of Philosophy - PhD, Robotics at University of MichiganHigh School, High School/Secondary Diplomas and Certificates, High School, High School/Secondary Diplomas and Certificates at Sir Winston Churchill High SchoolBachelor’s Degree, Engineering Science Robotics, Bachelor’s Degree, Engineering Science Robotics at University of Toronto
Github Skills (15)
mpc10
fpga10
verilog10
pytorch10
machine-learning10
python10
model-predictive-control10
eda10
cad10
flask-ask9
flask9
neural-network9
benchmarking9
benchmark9
numerical-optimization7
Programming languages (9)
Github contributions (5)
verilog-to-routing/vtr-verilog-to-routing
Jul 2015 - Sep 2015
Verilog to Routing -- Open Source CAD Flow for FPGA Research
Role in this project:
Back-end Developer
Back-end DeveloperContributions:80 commits, 2 PRs, 30 pushes in 2 months
Contributions summary:Sheng made several commits focused on improving the functionality of the VTR (Verilog-to-Routing) flow, specifically by adding and enhancing server-side logging, including logging to files. They also modified the server-side database tracking and scripts to aid benchmarking. These changes suggest the user was involved in enhancing the back-end functionality of the CAD flow's components, likely for better monitoring and data management.
vtrcadedaplacementsynthesis
locuslab/mpc.pytorch
Jan 2020 - Jan 2020
A fast and differentiable model predictive control (MPC) solver for PyTorch.
Role in this project:ML Engineer
ML EngineerContributions:6 commits, 1 PR, 4 comments in 1 day
Contributions summary:Sheng primarily contributes to example implementations within the repository, showcasing their expertise in applying model predictive control (MPC) techniques. Their contributions involve adding and modifying examples to solve control problems using PyTorch. Key additions include an example using a neural network-based dynamics model for a pendulum, and they have focused on tasks like integrating dynamics, defining cost functions, and integrating with the gym environment.
pytorchdifferentiablempcdeep-learningmodel-predictive-control
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.