Technical Team Lead Dexterous Manipulation
Role details
Job location
Tech stack
Job description
Our AI Research division is looking for a dedicated Dexterous Manipulation lead to move robots beyond pick-and-place toward tactile, contact-rich, human-level interaction. As Technical Team Lead, you will define the roadmap, guide a cross-functional team, and turn advanced imitation and reinforcement learning research into deployable robotic skills. Your mission: extend the existing large robot model research to high-DoF hands for a significant increase in fine manipulation capabilities., * Leadership & Architecture: Define the team's research roadmap and architecture for contact-rich manipulation; bridge across Robot Teaching, Foundation Models, and Hardware R&D to align on data, control, and execution needs.
- Development & Experimentation: Design manipulation policies combining visual, tactile, and force-torque feedback; supervise data pipelines using teleoperation, motion capture, and simulation (Isaac Sim, Omniverse, MuJoCo).
- Optimization & Integration: Lead on-site experiments in industrial use-cases and sim-to-real transfer, ensuring impactful results on real hardware for real-world tasks.
- Collaboration & Mentorship: Partner with external research and data collection collaborators; mentor engineers and researchers to build a pragmatic, reproducible, and impact-driven culture., * Infrastructure: Experience with GPU-based distributed training, cloud ML platforms (AWS, GCP, Azure), and modern CI/CD workflows.
- Sensing & Systems: Familiarity with tactile sensors, real-time control, or hardware-software co-design.
- Research Engagement: Publications or collaborations in dexterous manipulation, imitation learning, or reinforcement learning.
Requirements
Do you have a Master's degree?, * Education & Experience: Master's or PhD in Robotics, Computer Science, or related field; 5+ years in robotics R&D with focus on robotic hands, manipulation, machine learning, with prior leadership of projects or teams.
- Technical Expertise: Proficiency in Python/C++, PyTorch/TensorFlow, and ROS2; experience with simulation frameworks (Isaac Sim, Omniverse, or MuJoCo).
- Learning & Control: Deep understanding of Imitation Learning, Behavior Cloning, Reinforcement Learning, kinematics, and force-torque/trajectory control for high-DOF systems.
- Integration & Communication: Proven ability to transfer policies from simulation to hardware and communicate technical concepts clearly across disciplines.