Pubblicato 2026-04-22
Responsible for developing high-fidelity simulation environments and scalable digital twins to accelerate humanoid robotics development and deployment. The role focuses on building modular assets (robots, sensors, scenes) and enabling efficient simulation pipelines for reinforcement learning, testing, and validation. It bridges simulation and real-world systems by supporting sim-to-real transfer, synthetic data generation, and cross-team integration with control, AI, and mechanical engineering. Emphasis is placed on performance optimization, reproducibility, and creating structured environments suitable for parallel experimentation and training.
Responsibilities
Design, build and maintain physically realistic simulation environments and
robot models for humanoid tasks (manipulation, navigation, interaction)
Develop modular Digital Twin assets (robot, sensors, scenes) using Isaac Sim / USD
Enable rapid simulation cycles by:
Structuring environments for RL compatibility
Optimizing simulation performance for scalability and parallel training
Implement synthetic data generation pipelines (state, vision, teleoperation data)
Collaborate closely with multidisciplinary teams including: Motion and control
Engineers, Mechanical design team, AI / learning engineers
Contribute to sim-to-real alignment.
Requirements
Degree in Robotics, Automation, Computer Science or related field.
At least 2 years experience in robotics simulation, physics-based environments, or
game engines
Strong Python (C++ or similar languages is a plus)
Experience building environments and benchmarks using simulators such as Isaac
Sim , Mu Jo Co, Py Bullet, Unreal or Unity
Professional English level
Familiarity with:
Simulation pipelines (assets, scenes, sensors simulation)
Kinematics and basic dynamics
Basic rendering or physics concepts
Nice to Have
Experience with USD / Omniverse ecosystem
Exposure to synthetic data generation
Experience supporting RL training environments
