Acting Head of Reinforcement Learning Department
04/2026 – PresentLeading the RL engineering team and directing production deployment, while developing TACT-ful and CompliantWBC and mentoring 8 interns across locomotion policy design, sim-to-real transfer, and reward engineering.
Senior Reinforcement Learning Engineer
12/2025 – 03/2026Built and cross-validated multi-physics simulation environments (PhysX, MuJoCo, Newton/Warp) and high-throughput distributed GPU RL training pipelines, working with hardware and controls teams to keep policies compatible with real joint controllers, torque limits, and safety constraints.
R&D Software Engineer, Reinforcement Learning
06/2025 – 11/2025Trained RL policies for humanoid locomotion in Isaac Gym and Isaac Lab, using event-driven gait initialization and curriculum learning to reach stable walking, disturbance recovery, and adaptive balancing on hardware.
Research Assistant
05/2024 – 05/2025Generated 1M+ synthetic images in NVIDIA Omniverse and improved multi-view pose estimation by 40% through combined synthetic and real-data fine-tuning, enabling perception models accurate to within 30 cm in real warehouse environments.
Teaching Assistant, Computer Vision & Computational Photography
08/2024 – 12/2024Led lab sessions on multi-view geometry, camera calibration, and image-based reconstruction.
Earlier
Process Engineer, Power Control & Information Systems
01/2023 – 05/2023Built digital twins and ML models to optimize factory energy consumption.
Technology Specialist
07/2022 – 12/2022Researched depth estimation and built vision-based safety monitoring for construction sites.
Undergraduate Research Assistant
04/2021 – 06/2022Built structured-light 3D reconstruction and PointNet-based bin-picking segmentation pipelines.
Undergraduate Technical Intern
02/2021 – 08/2021Supported equipment qualification and data analysis for chipset production.
MSE in Robotics
Featured by Penn Engineering Graduate Admissions: “From Vietnam to Penn: Thanh Ly’s Journey as an International Student in Robotics” (Jan 2025)
BEng in Mechatronic Engineering, summa cum laude
Thesis: Integration of an Industrial Robot with a 3D Scanning Stereo Vision System Using Gray Code and Phase Shift
Infrastructure-Based Autonomous Forklift System
Infrastructure-mounted perception combining BEV global localization with FPV local planning to cut onboard sensing and compute needs, with multi-camera calibration and simulation for motion planning and collision checking.
Learning in Robotics, University of Pennsylvania
Trained RL policies in custom simulation environments for agile drone navigation under dynamic constraints, investigating sim-to-real policy robustness.
Interactive Computer Graphics, University of Pennsylvania
Real-time 3D voxel rendering engine in C++ with a custom OpenGL pipeline, chunk-based procedural terrain, and GLSL shaders for dynamic lighting, shadows, and physically-based materials.
Advanced Topics in Machine Perception, University of Pennsylvania
Visualized Stable Diffusion 3's internal attention structure using Normalized Cuts.
