SDE - Robot Manipulation & Learning, Frontier AI Robotics

We are seeking to hire research engineers for robot dexterous manipulation with the goal of developing dexterous manipulation systems at human-level performance. Our strategy focuses on a real2sim2real pipeline that learns from large scale human videos and leverage simulation and reinforcement learning to bridge the human-to-robot embodiment gap.

Key job responsibilities
• Learning dexterous manipulation: Design and implement robot learning algorithms for manipulation
• Leverage vision models to extract 3D hand-object manipulation information from real-world situations
• Develop large-scale real2sim2real pipeline for dexterous manipulation
• Multimodal Sensor Fusion: Integrate tactile sensing, proprioception and vision into the learning pipeline