Research Scientist, Atlas Behavior Learning

At Boston Dynamics, we’re pushing the boundaries of what robots can do—and we’re looking for curious, driven, and collaborative researchers to join us. As a Research Scientist on the Atlas Behavior Learning team, you’ll be part of a world-class group building groundbreaking whole-body manipulation capabilities for next-generation humanoids.

Whether you’re finishing your PhD or have a Master’s degree with hands-on experience in industry or research, this role offers a rare chance to apply cutting-edge ML to physical robots with full access to Atlas, our state-of-the-art humanoid platform.

How you will make an impact: 

• Design, train, and deploy reinforcement learning models to tackle challenging mobile and bimanual manipulation tasks

• Develop high-quality, production-ready code in Python and C++

• Test and iterate your models directly on real robot hardware

• Collaborate closely with expert engineers and researchers through design reviews and hands-on experimentation

• Contribute to a growing body of work shaping the future of humanoid robotics

We are looking for: 

• An MS or PhD in Computer Science, Robotics, Machine Learning, or a related field

• Experience training and deploying RL policies for complex behaviors—either in simulation or on real robots

• Familiarity with modern ML frameworks and tools (e.g., PyTorch, TensorFlow, Ray RLlib)

• A strong foundation in algorithm design, debugging, and performance optimization

Nice to have: 

• A PhD or equivalent research experience in reinforcement learning or robotic manipulation

• Experience working with real-world robotic systems

• Contributions to large software projects or open-source ML/robotics tools

Why join us? 

• Direct access to cutting-edge robots and the infrastructure to run large-scale experiments

• A collaborative, mission-driven team where your ideas have real impact

• The chance to help define what’s possible in real-world robotics

#LI-JM1