Robotics RL & Control Engineer – Whole-Body Locomotion

Walden Robotics, Inc.; Cambridge (MA); On-site; USD 140,000 - 190,000; Full time

Benefits offered by this job

Cash bonus

Equity

Company-subsidized insurance programs

401(k) with company match

Flexible PTO

Daily lunch

Job summary

Walden Robotics, Inc. is seeking a Member of Technical Staff – Engineer, RL and Control to advance learned controllers for humanoid robots, train policies, run simulations, and bring algorithms to real hardware.

You will collaborate with senior RL, controls and hardware engineers, grow in reinforcement learning, and own workstreams end-to-end from experimentation to deployment on real robots.

Qualifications

• Strong foundations in reinforcement learning, policy optimization, and sim-to-real.
• Solid knowledge of optimization-based control (trajectory optimization, MPC).
• Understanding of robot dynamics, kinematics, and control fundamentals.
• Experience bringing up controllers on physical robots.
• Proficient Python; C++ willingness to grow.

Responsibilities

• Train, tune, and evaluate reinforcement learning policies for whole-body control.
• Set up and run large-scale simulation experiments; analyze results.
• Contribute to sim-to-real pipeline with domain randomization and system identification.
• Bring controllers up on physical robots and debug behavior.
• Develop tools to evaluate controller performance in simulation and on hardware.
• Collaborate with RL, controls, and hardware engineers; own workstreams end-to-end.

Skills

Reinforcement learning

Model-based control

Robot dynamics & kinematics

Real robot experience

Python / C++ programming

Robotics simulation

Curiosity & ownership

Tools

Drake

MuJoCo

Python

C++

Job description

Walden Robotics, Inc. is seeking a Member of Technical Staff – Engineer, RL and Control to advance learned controllers for humanoid robots, train policies, run simulations, and bring algorithms to real hardware.

You will collaborate with senior RL, controls and hardware engineers, grow in reinforcement learning, and own workstreams end-to-end from experimentation to deployment on real robots.