Postdoctoral Researcher, Human Aware Interaction Learning

About the role

At Toyota Research Institute (TRI), we’re on a mission to improve the quality of human life by developing new tools and capabilities that amplify the human experience. We’ve built a world‑class team advancing the state of the art in AI, robotics, driving, and material sciences. The Opportunity: our team is looking for a motivated and capable postdoc to conduct research on the computational and machine learning aspects of human‑robot interaction, with interest in applying foundation models (e.g., VLAs, WAMs) to understand human‑robot collaboration and how it evolves as foundation models become central to robotic manipulation. We seek a solution‑driven attitude, close partnership with our research team, and the ability to turn ideas into impactful results. Relevant background includes robot foundation models, imitation/reinforcement learning, language‑conditioned trajectory planning/prediction, decision‑making under uncertainty, and human‑machine teaming/shared control. The Challenge: we are exploring novel approaches to efficiently learn how humans and robots collaborate, aiming to create AI approaches that challenge traditional interactions and enable new embodied interactive policies. The project focuses on a specific sub‑area toward publications in top‑tier conferences, suited for a postdoc seeking an additional research year before a faculty position or an experienced faculty member taking a year‑long sabbatical. The fellowship is based at our Cambridge, MA research center, where you will work with researchers on meaningful problems, access resources for ambitious projects and publications, and are encouraged to apply if you have relevant publications in these fields.

Responsibilities

• Perform research and publish on relevant topics in appropriate venues.
• Depending on the exact project outcome, publication target venues include, e.g., ICRA, CVPR, NeurIPS, and HRI.
• Publication emphasis is on how novel representations allow robots to learn to interact with humans towards long‑term shared autonomy in a data‑efficient, robust, and explainable way, with results in our teams’ problems and domain of interest.
• Exploration of both computational and cognitive phenomena, working with a team of researchers to create new approaches for understanding, predicting, and interacting with humans.
• Work from approach inception and ideation to validation of the developed approaches.

Qualifications

• Ph.D in related fields (by end date) – ML/AI, robotics, human‑centric AI, or computer vision.
• Some experience with robot systems is needed.
• Strong publications record in top ML/robotics venues.
• Specific areas: robot foundation models and their use for human interaction, ML and synthetic data for human‑collaborative embodied reasoning, imitation/reinforcement learning, and learning‑based shared control and planning for human interactions.
• Coding: Python, experience working in a team on joint scientific projects.
• Experience training deep learning models in e.g., PyTorch/TensorFlow.
• An ideal candidate can refine a topic and execute a scientific research plan in collaboration with other researchers.

Benefits

• Medical, dental, and vision insurance
• 401(k) eligibility
• Paid time off benefits (including vacation, sick time, and parental leave)
• Annual cash bonus structure

Pay

The pay range for this position at commencement of employment is expected to be between $136,800 and $196,650/year for Massachusetts‑based roles. Base pay offered will depend on multiple individualized factors, including, but not limited to, a candidate's experience, skills, job‑related knowledge, and market location.