Skip to content

Jobs

Amazon Robotics

Robot type
Warehouse and Logistics
Location
North ReadingMassachusettsUSA
Job type
Artificial Intelligence
Posted
Aug 31, 2026
Full-time

Principal Applied Scientist, Contact-Rich Manipulation, AR Manipulation HW & Systems

Job description

How should a robot dig a single item out of a cluttered bin, feel when it has made contact, and adjust in real time without crushing what it touches? Single- and dual-arm contact-rich manipulation at production scale is still being solved. We are looking for a Principal Applied Scientist to define the control architectures that our next-generation grasping systems will be built on.

The Manipulation Robotics team develops robotic workcells to pick, grasp, and move millions of items and packages across Amazon's fulfillment network. Reaching into clutter and grasping deformable and diverse items depends on solving contact: force-aware control, compliant behaviors, and the layering of classical…

Job requirements

  • Experience distilling informal customer requirements into problem definitions, dealing with ambiguity and competing objectives
  • PhD in Robotics, Machine Learning, Computer Science, Electrical Engineering, Mechanical Engineering, or a related technical field, or an equivalent body of published or deployed work in place of the PhD
  • Experience programming in Python and C++ (or a related language), including production-quality code deployed on real robotic hardware.
  • Demonstrated expertise in force, impedance, or admittance control for contact-rich manipulation.
  • Experience integrating learned policies with low-level force or impedance controllers, including the choice of action representation for contact-rich tasks.
  • Track record of setting the technical direction adopted by your organization, company, or research community.
  • 10+ years of industry or academic research experience
  • Hands-on experience training manipulation policies with reinforcement or imitation learning, and sim-to-real transfer of those policies onto real robotic hardware for contact-rich tasks.

Similar jobs