Research Scientist, Robotics Research

Toyota Research Institute•Cambridge, MA

2d•$158,400 - $227,700

About The Position

Toyota Research Institute (TRI) is seeking a researcher to help develop robots that can work reliably alongside people in real, unstructured environments. The robotics team brings together expertise in hardware, simulation, perception, controls, and machine learning, with success measured through fundamental capability development, open-source software, and publications. The Research Scientist will contribute to useful robots and general-purpose robot foundation models, developing robot behavior-learning methods from interactive embodied data and online data sources. The role includes building scalable, efficient, high-performance machine-learning and data pipelines, contributing to internal capabilities and external research publications, and collaborating with TRI researchers and partner laboratories at leading universities.

Requirements

PhD in computer science, machine learning, robotics, or a closely related field.
Experience training large models and deploying them on embodied systems, particularly for robotic manipulation.
Strong Python software development skills, familiarity with mixed C++/Python codebases, and a focus on clean, maintainable code.
Extensive practical machine-learning experience with a major framework such as PyTorch or TensorFlow.
Strong understanding of state-of-the-art robot learning, including generative models such as diffusion policy or flow matching, reinforcement learning, and/or world models.
Practical experience with robots and the system-integration challenges involved in research and deployment on physical hardware platforms.
Ability to move quickly and switch between rapid prototyping and robust implementation as needed.
Strong record of impact through first-author publications at top-tier machine-learning or robotics conferences, or meaningful contributions to successful industry initiatives.

Nice To Haves

Experience with data pipelines, model serving and optimization, cloud training, and dataset management.
Experience in robotics and machine-learning research or related projects in an industry setting.
Experience with robotic middleware such as ROS 2 and common communication methods and protocols.
Experience with modern machine-learning infrastructure pipelines, approaches, and tools.
Experience with VR-based teleoperation for real-time robot control.
Background or familiarity with motion control and actuation, whole-body control, reinforcement learning, robot teleoperation methods, communication protocols, research robotic arms or systems, visual perception and depth sensors, machine learning, robotic simulation, force and tactile sensing systems, or haptic interfaces.

Responsibilities

Work as part of a closely knit research team building useful robots and general-purpose robot foundation models.
Implement, extend, and create state-of-the-art methods for robot behavior learning using interactive embodied data and online data sources.
Design and implement high-performance machine-learning pipelines and optimize data and learning stacks for scalability, efficiency, and performance.
Contribute to rapid team progress through development of internal capabilities and high-impact external publications.
Collaborate with internal research scientists and partner laboratories at MIT, Stanford, Berkeley, CMU, Columbia, and Princeton to advance research.