Software Engineer - Dexterous Manipulation

📋 Description

• Implement and tune control algorithms for multi-fingered hands—grasping, in-hand manipulation, and
• Develop and maintain manipulation software with low-latency, reliable execution inside the
• Translate RL/imitation policies and teleoperation retargeting into production-grade C++/Python.
• Build and refine sim-to-real pipelines—hand models, domain randomization, validation in
• Deploy and debug manipulation capabilities on physical robots; diagnose sensor-noise, latency, and
• Collaborate with hardware and systems engineers to improve hand performance and sensor/actuator

🎯 Requirements

• Dexterous manipulation experience: multi-fingered grasping, in-hand manipulation, high-DOF hand
• Reinforcement learning for robotic control—reward design, training, debugging; imitation learning
• Strong Python and working C++ for real-time robotic software.
• Hands-on experience training and validating policies in physics simulators (IsaacSim, MuJoCo
• Robotics fundamentals: kinematics, dynamics, Jacobian-based control.
• Track record of deploying algorithms on physical hardware, not just in simulation.

🎁 Benefits