Robotics Control Engineer – Humanoid Robots

About the Role

Hands-on Robotics Control Engineer to design, implement, and optimize control algorithms that enable humanoid robots to balance, walk, manipulate objects, and operate reliably in real-world environments. You will work directly with physical robots, actuators, sensors, embedded systems, and simulation platforms, collaborating with robotics, VLA, AI, and hardware teams.

Key Responsibilities

Humanoid Robot Control Systems

• Design and implement joint-level, task-space, and whole-body controllers for humanoid platforms.
• Develop position, velocity, torque, and impedance control strategies.
• Integrate controllers with robot hardware and improve stability, responsiveness, and reliability.

Whole-Body Control & Robot Dynamics

• Implement inverse kinematics, inverse dynamics, and optimization-based control.
• Use quadratic programming and other methods to manage contact forces, joint constraints, and actuator limits.
• Coordinate arms, legs, torso, and hands during complex movements.

Locomotion, Balance & Motion Planning

• Develop walking, standing, gait generation, footstep planning, and trajectory optimization.
• Implement Model Predictive Control (MPC), posture stabilization, disturbance rejection, and recovery strategies.
• Improve dynamic locomotion across changing surfaces and operating conditions.

Robotic Manipulation & Motion Control

• Develop trajectory tracking, force control, compliant manipulation, and bimanual coordination.
• Improve grasping, object handling, and contact-rich task execution.
• Connect VLA or learned policies to reliable low-level robotic controllers.

Real-Time Control & Hardware Integration

• Integrate motors, actuators, encoders, IMUs, and force/torque sensors.
• Optimize control frequency, latency, hardware communication, and computational efficiency.
• Support calibration, commissioning, joint limits, and emergency-stop safety mechanisms.

Simulation & Sim-to-Real

• Build and validate control systems in Isaac Sim, Isaac Lab, MuJoCo, Gazebo, or similar tools.
• Validate dynamic models and control behavior before physical deployment.
• Reduce sim-to-real gaps and establish repeatable testing procedures.

Learning-Based Control

• Integrate reinforcement learning and imitation learning policies with model-based controllers.
• Work with AI researchers on learned locomotion and manipulation.
• Evaluate robustness and implement safety constraints and fallback behaviors.

State Estimation & Sensor Integration

• Integrate IMUs, joint encoders, force sensors, and proprioceptive feedback.
• Develop sensor fusion for orientation, velocity, position, and contact-state estimation.
• Investigate calibration, noise, and estimation instability.

Testing, Validation & Collaboration

• Benchmark stability, movement accuracy, latency, repeatability, and task completion.
• Analyze physical robot experiments, troubleshoot failures, and document improvements.
• Collaborate with hardware engineers, VLA engineers, AI researchers, and robotics operations.

Requirements

• 5+ years of professional experience in robotics control, motion control, autonomous systems, or related engineering.
• Bachelor’s or Master’s degree in Robotics, Mechanical Engineering, Electrical Engineering, Mechatronics, Computer Science, or related field.
• Advanced proficiency in C++ and Python; experience with ROS/ROS2 and Linux.
• Strong knowledge of robot kinematics, dynamics, control theory, PID, impedance control, inverse kinematics, and trajectory tracking.
• Knowledge of MPC, optimization-based control, or whole-body control.
• Experience with multi-DOF robotic systems, actuators, sensors, motor controllers, and hardware interfaces.
• Experience with robotics simulation, real-time systems, and deploying algorithms on physical robots.
• Strong debugging, analytical, problem-solving, and cross-functional communication skills.

Preferred Qualifications

• Direct experience with humanoid or bipedal robots, dynamic walking, and balance.
• Experience with QP-based control, trajectory optimization, rigid-body dynamics, and contact mechanics.
• Familiarity with Pinocchio, Drake, Crocoddyl, MuJoCo, NVIDIA Isaac Sim or Isaac Lab.
• Experience with real-time Linux, EtherCAT, CAN, torque-controlled actuators, and high-frequency control loops.
• Experience with dexterous manipulation, bimanual control, reinforcement learning, and learned policy integration.
• Experience with state estimation, sensor fusion, research publications, or open-source robotics contributions.

Key Performance Indicators

• Stability and reliability of humanoid control systems; walking and balance performance.
• Accuracy and repeatability of motion, manipulation, and task execution.
• Control latency, execution errors, and robustness across operating conditions.
• Successful physical robot integration and simulation-to-real deployment.
• Achievement of engineering milestones, test coverage, and documentation quality.

What Success Looks Like in the First 90 Days

Days 1–30 | Assessment & Familiarization

• Review humanoid hardware, control software, sensors, and existing locomotion/manipulation capabilities.
• Establish baseline benchmarks and identify priority improvements.

Days 31–60 | Development & Integration

• Implement and simulate selected control improvements.
• Integrate controllers on physical robots and demonstrate measurable gains.

Days 61–90 | Validation & Optimization

• Validate balance, locomotion, and manipulation performance on physical robots.
• Optimize stability and latency; present results and a six-month development roadmap.