MTS - Research (Robotics Simulation Data)
Core
Design and build the physics-solving engine for a synthetic data generation platform to support AI model training.
Role type
Senior IC applied scientist (robotics simulation)
Builds
High-fidelity physics simulation stack and synthetic training datasets
Domain
Robotics, Physics Simulation, AI/ML
Deliverable
production ML models
Required skills
Physics engine development, GPU programming (CUDA), C++, Python, numerical methods, contact dynamics, deformable simulation, fluid simulation, differentiable simulation, performance profiling, procedural scene generation
Preferred skills
Publications in numerical computing/graphics, open-source contributions, shipped commercial systems in physics simulation
Technologies
CUDA, C++, Python
Responsibilities
Develop advanced physics solvers for high-degree-of-freedom interactions and complex contact dynamics; Design simulation environments and domain randomization strategies for high-utility training data; Scale environment variations and dynamic multi-agent interactions across billions of simulation steps; Profile and optimize simulation execution in multi-GPU, parallelized pipelines; Shape long-term roadmap for high-fidelity physical modeling and synthetic data engine architecture
