Resume
Summary
Zimu (Zim) Gong
Robotics research engineer at Meta FAIR focused on VLA models, reinforcement learning, and deploying machine learning models to robots.
- San Francisco Bay Area
- zimgong@umich.edu
Education
Master of Science, Computer Science
Aug 2024 - May 2026
University of Michigan, Ann Arbor, MI
Bachelor of Science, Electrical and Computer Engineering
Sep 2020 - Aug 2024
UM-SJTU Joint Institute, Shanghai Jiao Tong University, Shanghai, China
Study Abroad, Electrical and Computer Engineering
Jan 2023 - May 2023
Cornell University, Ithaca, NY
Professional Experience
Research Engineer
Jun 2026 - Present
Meta · FAIR · Menlo Park, CA
- Researching robotics and physical AI as part of FAIR.
- Working on VLA models, reinforcement learning, and robot intelligence.
Research Engineer / Research Intern
Jan 2026 - Jun 2026
Assured Robot Intelligence · San Diego, CA
- Founding member of ARI, which was acquired by Meta.
- Built research systems for industry-grade physical AI and robotics applications.
Research Intern
May 2025 - Aug 2025
Waabi · A&A Behaviors Team · San Francisco, CA
- Worked on traffic simulation and safety-critical scenario generation.
- Developed traffic models with flow-matching VAE methods.
- Contributed to work accepted at ICRA 2026.
Autonomous Division Member
Aug 2024 - May 2025
MRacing · Ann Arbor, MI
- Worked on mapping and localization for autonomous FSAE.
- Designed and implemented an EKF-SLAM algorithm in ROS for autocross events.
Research Intern
May 2024 - Aug 2024
Lightwheel
- Participated in the development of an in-house simulator supporting Unreal Engine and 3DGS hybrid rendering.
- Trained and evaluated end-to-end autonomous driving models.
- Worked on simulation, imitation learning, and reinforcement learning for manipulation tasks.
Drive Sim Intern
Jun 2023 - Aug 2023
Lightwheel
- Developed a data visualization system for custom synthetic autonomous vehicle simulation data i.e. (bounding boxes, maps, diagnostics) with Python Foxglove framework.
- Developed a synthetic data generation system with Unreal Engine that features keyboard control, customizable sensor modelling for cameras and lidars and smart agent system that utilizes a built-in navigation mechanism in UE.
- Developed and maintained a Python library that contains sensor models and 3D transformation methods.
Publication
Conditional Flow-VAE for Safety-Critical Traffic Scenario Generation
ICRA 2026 · May 2026
Accepted to the IEEE International Conference on Robotics and Automation.