About the job
Agility Robotics creates humanoid robots designed to assist teams in warehouses, manufacturing, and distribution centers. By automating physically demanding and repetitive tasks, these robots help people focus on more strategic work. Agility Robotics has a proven record of safety and real-world deployment, with a mission to broaden the role of automation in supporting human labor.
Role overview
The Senior AI Research Engineer joins the AI innovation team to develop and deploy foundational models and end-to-end policies for humanoid robots. This position centers on advancing imitation learning and reinforcement learning, enabling robots to perform a variety of real-world activities. The work is hands-on, involving direct experimentation with robots and close collaboration with a multidisciplinary team to transition new methods from research into production.
Key responsibilities
- Design, train, and deploy advanced policies for locomotion, manipulation, and interaction in dynamic environments.
- Build reinforcement learning infrastructure, including scalable training pipelines and evaluation tools.
- Create simulation environments and tasks to support the training and deployment of control policies.
- Research, develop, and test new imitation learning techniques.
- Collaborate with Robotics Software and AI engineering teams to ensure policies are ready for production use.
Requirements
- Minimum of 3 years of experience developing and deploying learning-from-demonstration methods.
- Strong programming skills in Python and deep experience with frameworks such as PyTorch.
- Background in reinforcement learning and robotics is highly valued.
Location
This hybrid position can be based at any Agility Robotics office: Fremont, CA; Salem, OR; or Pittsburgh, PA.
