Senior Simulation Engineer - Robotics Manipulation Simulation

London
- Fully onsite Full time SC-E

Join a forward-thinking engineering team to develop and maintain simulation environments for dexterous manipulation tasks in industrial, service, and home settings. This role focuses on creating realistic physics-based simulations and reinforcement learning policy training to bridge the sim-to-real gap. Collaborate with vendors to integrate 3D assets, optimize simulation performance, and enhance learning infrastructure to support rapid development cycles in robotics simulation.

• Design and implement simulation gym environments for manipulation tasks across multiple domains
• Define observation spaces, action spaces, and reward functions for reinforcement learning
• Analyze and mitigate reward hacking to ensure robust learned behaviors
• Tune physics parameters and improve asset fidelity to reduce sim-to-real transfer gaps
• Collaborate with vendors to validate and integrate high-quality 3D simulation assets
• Ensure accurate physics setup including dynamics, friction, mass properties, and joint limits
• Identify and resolve simulation performance bottlenecks for efficient training
• Enhance simulation evaluation and reinforcement learning infrastructure for scalability
• 3+ years in simulation environment development or game-engine-based interactive systems
• Proficient with physics simulators such as Isaac Sim, Mu Jo Co , PyBullet, or equivalent game engines like Unreal or Unity
• Experience with large-scale GPU-accelerated simulation and parallel processing
• Strong Python programming skills with ability to profile and debug physics issues
• Familiarity with modern software engineering practices and clear experiment documentation
• Desirable: experience with robotics manipulation, reward function design, or reinforcement learning training
• Knowledge of sim-to-real transfer techniques such as domain randomization or system identification
• Publications or open-source contributions in robotics or reinforcement learning domains

Confirm any vetting requirements with your consultant. An absent badge does not mean vetting is unnecessary.

London
- Fully onsite, Full time. Confirm employment terms and any clearance requirements with your consultant.

simulation robotics reinforcement learning physics simulation Python GPU acceleration RL policy training manipulation