Applied Scientist - Robotics Controls

Amazon•North Reading, MA

About The Position

At Amazon Robotics, we design advanced robotic systems capable of intelligent perception, learning, and action alongside humans, all on a large scale. Our goal is to develop robots that increase productivity and efficiency at the Amazon fulfillment centers while ensuring the safety of workers. We are seeking an Applied Scientist to develop innovative, scalable solutions in feedback control and state estimation for robotic systems, with a focus on contact-rich manipulation tasks. In this role, you will formulate physics-based models of robotic systems, perform analytical and numerical studies, and design control and estimation algorithms that integrate fundamental principles with data-driven techniques. You will collaborate with a world-class team of experts in perception, machine learning, motion planning, and feedback controls to innovate and develop solutions for complex real-world problems. As part of your work, you will investigate applicable academic and industry research to develop, implement, and test solutions that support product features. You will also design and validate production designs. To succeed in this role, you should demonstrate a strong working knowledge of physical systems, a desire to learn from new challenges, and the problem-solving and communication skills to work within a highly interactive and experienced team. Candidates must show a hands-on passion for their work and the ability to communicate their ideas and concepts both verbally and visually.

Requirements

PhD, or Master's degree and 4+ years of science, technology, engineering or related field experience
Experience programming in Java, C++, Python or related language
Experience implementing algorithms using both toolkits and self-developed code
Demonstrated ability to innovate and deliver robust, scalable solutions for complex robotics problems in production environments.
Deep expertise in developing force-based and impedance/control strategies for contact-rich manipulation.
Experience with state-estimation techniques, including classical observers, Kalman Filters, EKF/UKF, and Particle Filters.
Experience modeling robotic or physical systems for simulation, as well as characterizing and validating system dynamics on hardware.

Responsibilities

Research, design, implement, and evaluate feedback control, estimation, and motion-planning algorithms, ensuring effective integration with perception, manipulation, and system-level components.
Develop experiments, simulations, and hardware prototypes to validate control algorithms, and optimization techniques in contact-rich manipulation and other challenging scenarios.
Collaborate with software engineering teams to enable scalable, real-time, and maintainable implementations of algorithms in production systems.
Partner with cross-functional teams across hardware, systems engineering, science, and operations to transition algorithms from early prototyping to robust, production-ready solutions.
Engage with stakeholders at all levels to iterate on system design, define requirements, and drive integration of control and estimation capabilities into Amazon Robotics platforms.