RESPONSIBILITIES：
• Design, implement, test, and deploy motion planning algorithms for high-DOF manipulators, with emphasis on contact-rich and compliant manipulation tasks such as assembly and surface treatment
• Co-own the software interface between the motion planning stack and the whole body control module; define and maintain shared state representations, constraints, and control handoff protocols
• Develop planners that reason over contact modes, contact sequencing, and force/torque constraints — not just collision-free path finding
• Benchmark and evaluate planners in simulation and on real hardware across contact-rich scenarios; own reliability and task-success metrics
• Collaborate across perception, control, and hardware teams to translate physical task requirements into well-defined planning problems
• Drive software quality through code review, testing standards, and continuous improvement of engineering best practices
• Manage and communicate development schedules and milestones
REQUIREMENTS：
• PhD or MS in Robotics, Mechanical Engineering, Computer Science, or a related field — or equivalent industry experience
• 3+ years of hands-on experience with robotic systems software engineering
• Proficiency in C++ and/or Python; demonstrated experience deploying motion planning software on real robots and simulators
• Strong theoretical and practical understanding of motion planning for high-DOF manipulators, including planners that operate under contact and force constraints
• Solid grounding in robot kinematics and dynamics forward/inverse kinematics, Jacobian methods, rigid body dynamics, and force/torque reasoning
• Ability to work independently, take ownership, and continuously raise the bar on engineering standards
PREFERRED SKILLS
Strong candidates will have experience in one or more of the following areas:
• Experience with whole body control, impedance control, or admittance control for compliant manipulation
• Familiarity with contact mechanics and hybrid force-motion control e.g. force-controlled insertion, peg-in-hole, surface following, deburring, or polishing tasks
• Experience with trajectory generation for robot manipulators, including time-optimal parameterization and smooth Cartesian trajectory design e.g. time-optimal path parameterization (TOPP), C² continuous Cartesian trajectories, jerk-limited motion profiles, spline-based or Bézier representations; awareness of how trajectory smoothness affects contact stability and surface quality
• Background in numerical optimization and optimal control e.g. trajectory optimization, MPC, QP solvers — especially with contact constraints
• Experience applying reinforcement learning to contact-rich or high-DOF manipulation e.g. model-free / model-based RL, sim-to-real transfer, policy learning for dexterous tasks
• Experience with ROS / ROS2 in multi-process, real-time robotic systems
• Familiarity with PyTorch / CUDA for scientific computing or learning-based planners
• Proven ability to pick up a new knowledge domain and deliver production-quality code
