Principal AI Research Engineer - RL

1 year ago

San Francisco, CA, USA or New York, NY, USAStaff+

Responsibilities

Develop robust on-policy reinforcement-learning policies for real-world robots.
Re-implement and improve core reinforcement-learning algorithms such as SAC and DDPG.
Debug unstable gradients and tune reinforcement-learning hyperparameters.
Contribute intellectual advances to sample-efficient reinforcement-learning algorithms such as DreamerV3 and MuZero.
Ship reinforcement-learning systems that learn on physical robot hardware and improve deployment reliability.

Experience re-implementing core RL algorithms such as SAC and DDPG from scratch.
Ability to debug unstable gradients and tune hyperparameters correctly.
Meaningful intellectual contributions to sample-efficient RL algorithms such as DreamerV3 and MuZero.
Experience shipping on-policy RL on hardware that learns in the real world, such as quadruped walking or drone racing.
A portfolio or significant project work is encouraged.

High ownership and high equity in a small startup.
Opportunity to help shape the product and build a robotics company from the ground up.
Work with in-house hardware, proprietary robotics data, and real-world robot deployments.

11-50 employees