Robotics Research Intern, Manipulation

Toyota Research Institute•Los Altos, CA

2d•Hybrid

About The Position

Toyota Research Institute (TRI) is seeking a Robotics Research Intern to investigate agentic approaches to robotic manipulation. The Robotics Division develops technologies that help robots operate effectively in complex, real-world environments. This project explores how foundation-model agents can understand manipulation tasks, reason about physical environments, and orchestrate reusable robotic capabilities. The research combines vision-language models, robotic perception, motion planning, manipulation, and robot learning, with hands-on experimentation on physical robotic platforms. The intern will work closely with researchers and engineers to connect foundation-model reasoning with physical robot capabilities. Depending on their background and research interests, the project may focus on spatial reasoning, robotic skill orchestration, execution monitoring, failure recovery, or systematic evaluation. The internship offers experience with cutting-edge AI models and real robotic systems, and contributes to research on generalizable and reliable robot autonomy.

Requirements

Currently pursuing a Master's or PhD degree in Robotics, Computer Science, Electrical Engineering, Mechanical Engineering, or a related field.
Strong programming skills in C++ and/or Python, with experience developing, testing, or debugging technical software.
Solid understanding of robotics fundamentals, including one or more of robot kinematics, dynamics, motion planning, perception, or control.
Experience with robotics-related projects through research, coursework, or personal projects, using physical hardware and/or simulation.
Ability to formulate technical problems, design experiments, and systematically analyze results.
Interest in robotic manipulation, autonomous systems, and connecting high-level reasoning with physical robot capabilities.

Nice To Haves

Hands-on experience with robotic manipulators, motion planning frameworks, or robot simulation environments.
Familiarity with 3D geometry, computer vision, grasp planning, or contact-rich manipulation.
Experience with ROS/ROS 2 or robotics software frameworks.
Experience with vision-language models, LLM agents, or autonomous planning systems.
Familiarity with robot learning methods such as imitation learning, reinforcement learning, or vision-language-action models.
Research experience demonstrated through publications, open-source contributions, or substantial technical projects.

Responsibilities

Develop and evaluate agentic approaches that use vision-language models to reason about and execute robotic manipulation tasks.
Integrate foundation-model agents with reusable robotic capabilities, including perception, motion planning, grasping, and manipulation.
Investigate three-dimensional spatial reasoning, robot action feasibility, execution monitoring, and failure recovery.
Design and conduct experiments on physical robotic platforms or in simulation.
Analyze system performance, characterize failure modes, and evaluate generalization across tasks and environments.
Communicate research findings through technical presentations, demonstrations, reports, and potential publications.