Staff Reinforcement Learning Engineer (Robotics)

TL;DR

Staff Reinforcement Learning Engineer (Robotics): Developing, training, and deploying advanced reinforcement learning algorithms for whole-body control of humanoid robots with an accent on sim-to-real gap reduction and policy performance optimization. Focus on defining robust control metrics, hardening the control stack, and leading complex technical projects in a high-collaboration environment.

Location: Must be based in San Jose, CA (5 days/week in-office)

Salary: $150,000–$250,000

Company

An AI robotics company building autonomous, general-purpose humanoid robots with human-level intelligence for home and commercial use.

What you will do

• Develop, train, and deploy RL algorithms for whole-body robot control.
• Determine optimal observations, actions, and model architectures.
• Identify and bridge sim-to-real performance gaps.
• Define, test, and evaluate metrics for learned policies.
• Harden the control stack to ensure system robustness.
• Lead complex controls projects and mentor junior team members.

Requirements

• Strong background in dynamics and control, specifically with legged robots.
• Experience with RL algorithms for robotics (PPO, SAC).
• Expertise in tuning hyperparameters and cost functions for RL.
• Familiarity with domain randomization, curriculum learning, and reward shaping.
• Experience leading technical projects and mentoring engineers.

Nice to have

• Experience with behavior cloning techniques such as distillation.