Senior RL Engineer - Humanoid Robotics (Equity)

Humanoid; United Kingdom; On-site; GBP 90,000 - 150,000; Full time

Benefits offered by this job

Private healthcare

Equity included

Pension plan

Office meals

Job summary

Humanoid is seeking a Senior or Staff Reinforcement Learning Engineer to develop learning-based control policies for humanoid robots. You will design and train reinforcement learning policies that enable dynamic locomotion and loco-manipulation behaviors on real robots, building scalable training pipelines and improving sim-to-real transfer for reliable hardware deployment.

You will work with controls and robotics engineers to integrate learned policies into the robot control stack, performing

Qualifications

• MS or PhD in Robotics, Machine Learning, Computer Science, or related field.
• Strong experience with reinforcement learning (e.g., PPO, SAC, offline RL).
• Experience applying RL to robotics or physical systems.
• Experience deploying learned policies on real robotic systems.
• Experience with physics-based simulation environments (e.g., Isaac Lab, MuJoCo).
• Strong programming skills in Python and/or C++.

Responsibilities

• Design and train reinforcement learning policies for humanoid robot control.
• Build scalable simulation and training pipelines (e.g., Isaac Lab, MuJoCo).
• Design reward functions, observation spaces, and curricula for complex behaviors.
• Improve robustness and sim-to-real transfer of learned policies.
• Deploy and evaluate policies on real robotic systems.
• Integrate policies into the control stack.

Job description

Humanoid is seeking a Senior or Staff Reinforcement Learning Engineer to develop learning-based control policies for humanoid robots. You will design and train reinforcement learning policies that enable dynamic locomotion and loco-manipulation behaviors on real robots, building scalable training pipelines and improving sim-to-real transfer for reliable hardware deployment.

You will work with controls and robotics engineers to integrate learned policies into the robot control stack, performing