Atlas Reinforcement Learning Intern

Role Overview

As an Atlas Reinforcement Learning Intern, you will join a world-class team of engineers and scientists focused on creating robust perception-driven manipulation policies for Atlas. You’ll work on a research project, advised by a senior research scientist, to deliver results onto hardware. Responsibilities include:

• Participating in project ideation
• Implementing ideas in the existing codebase
• Deploying policies on hardware with the Atlas team
• Measuring progress using manipulation benchmarks

Impact & Responsibilities

• Apply machine learning creatively to solve real-world bimanual manipulation problems
• Write high-quality, documented, tested, maintainable Python and C++ code
• Lead design reviews and collaborate with the Atlas software team
• Deploy and debug code on Atlas, Spot, and other BD robots

Qualifications

• Actively pursuing a PhD in computer science, machine learning, robotics, or a related field
• Prior experience training and deploying learned models to create complex behaviors in robots or simulated characters
• Experience with common ML tools, architectures, and workflow
• Strong analytical and debug skills

Nice to Have

• Experience working with and contributing to large codebases
• Experience working with robot hardware