Principal Research Scientist - Planning & Learning, Amazon Robotics

Amazon - North Reading, MA

Full-time - Principal

North Reading, MA

Sporting Goods, Hobby, Musical Instrument, Book, and Miscellaneous Retailers

About the position

Amazon Robotics is seeking a Principal Research Scientist to develop state-of-the-art algorithms to control the fleet of 750,000 mobile robots in Amazon's order fulfillment network. In this role, you will have the opportunity to solve complex planning and coordination problems and see your solutions come to life in Amazon's warehouses. You will own the science of planning and scheduling fleet activities to achieve system-level objectives across a broad range of building types and workflows. Your work will involve improving the algorithms that control today's global robotic fleet to unlock new opportunities for Amazon's evolving network, as well as developing new capabilities from the ground up. You will explore how to leverage the power of our robotic mobility foundation model to predict the motion of the fleet, enabling new ways of making decisions at scale. This position offers the chance to have a broad impact on the next wave of robotics, influencing the design of future warehouses and the integration of new robotic technologies. You will collaborate with brilliant colleagues in a culture that values ingenuity, teamwork, and delivering impact. Our multi-disciplinary science team includes scientists with backgrounds in simulation, planning and scheduling, grasping and manipulation, machine learning, and operations research. We develop novel planning algorithms and machine learning methods and apply them to real-world robotic warehouses. This includes planning and coordinating the paths of thousands of robots, dynamic allocation and scheduling of tasks to thousands of robots, learning how to adapt system behavior to varying operating conditions, and co-designing robotic logistics processes and the algorithms to optimize them. Our team also serves as a hub to foster innovation and support scientists across Amazon Robotics, coordinating research engagements with academia, such as the Robotics section of the Amazon Research Awards.

Responsibilities

• Develop state-of-the-art algorithms for controlling mobile robots in Amazon's fulfillment network.
• Own the science of planning and scheduling fleet activities to achieve system-level objectives.
• Improve existing algorithms that control the global robotic fleet.
• Develop new capabilities from the ground up for robotic operations.
• Leverage the robotic mobility foundation model to predict fleet motion.
• Influence the design of future warehouses and integration of new robotic technologies.
• Collaborate with a multi-disciplinary team to foster innovation and support research engagements with academia.

Requirements

• PhD degree in Robotics, Computer Science, or a related technical field.
• 10+ years of relevant, broad research experience after obtaining a PhD.
• Experience programming in C++, Java, Python, or related languages.
• Experience using state-of-the-art planning and machine learning methods.
• Strong publication record with contributions to leading conferences and journals.
• Extensive experience developing and deploying algorithms for planning and control of real-world applications.
• Deep knowledge of adjacent fields, including optimization, heuristic search, and deep learning.
• Comfort working in a fast-paced development environment.
• Ability to identify unsolved problems and creatively invent solutions.
• Proven track record of innovation.