About the job
We’re seeking an AI Intern to help our humanoid robot, Digit, navigate and interact with the world. You’ll be in the driver’s seat of an applied research project, moving past theory to see your code manifest in motion on hardware. You’ll work alongside some of the brightest minds in the industry, utilizing a fleet of Digits as your playground.
Responsibilities
Lead a research project in robot learning, such as reinforcement learning for whole-body control or learning loco-manipulation skills from demonstration.
Design and run experiments to evaluate learning-based robot behaviors in simulation and on hardware.
Improve data collection, model training, and evaluation frameworks.
Dive into robot logs and experimental data to diagnose failures and push the boundaries of what Digit can do.
Present your findings to a world-class team of engineers and researchers.
Qualifications
Minimum
Currently pursuing a PhD in Computer Science, Machine Learning, Robotics, or a related field.
Strong skills in Python and modern ML frameworks. (PyTorch, JAX, etc.).
Excited to integrate, run, and evaluate your learned policies on the robot.
Strong analytical and debugging skills.
Ability to work independently on open-ended technical problems.
Preferred
Expertise in one or more of the following areas:
Reinforcement learning
Learning from demonstration
Multimodal learning (vision, proprioception, forces, etc)
Sim-2-real
Experience training machine learning models on real or simulated datasets.
Experience working with robot hardware.
Experience analyzing experimental data and communicating results.