Research Intern: Collaborative Multi-Agent Intelligence - Honda Research Institute USA

Research Intern: Collaborative Multi-Agent Intelligence

Your application is being processed

Research Intern: Collaborative Multi-Agent Intelligence

Job Number: P25INT-08
​Honda Research Institute USA (HRI-US) is seeking a research intern to contribute to foundational research focusing on how intelligent agents can reason about themselves, infer others’ beliefs and intentions, and coordinate effectively in dynamic, multi-agent settings. This project integrates metacognition (self-reflection and uncertainty reasoning), Theory of Mind (ToM) (reasoning about others’ goals and beliefs), and collective reasoning (shared alignment and adaptation across agents). The intern will help design, prototype, and evaluate algorithms and representations that enable self-aware and interdependent agents — providing the groundwork for future systems capable of collaborative cognition and team-level intelligence.
San Jose, CA

 

Key Responsibilities

 
  • Design and evaluate multi-agent reasoning mechanisms that combine self–other belief modeling, inverse planning, and metacognitive control for uncertainty assessment and self-correction.
  • Develop and test collaborative reasoning frameworks using Vision-Language-Action (VLA) models to connect perception, language, and decision-making in multi-agent settings.
  • If required, develop or extend benchmark tasks and datasets (e.g., using VirtualHome, PARTNR) to evaluate collective reasoning and coordination in language-based or multimodal environments.
  • Collaborate with researchers across AI, cognitive science, and related disciplines.
  • Contribute to a portfolio of patents, academic publications, and prototypes to demonstrate research value.

 

Minimum Qualifications

 
  • Currently enrolled as a Ph.D. student in Computer Science, Cognitive Science, Robotics, Machine Learning or related fields at a reputed university. (exceptional M.S. candidates with a minimum of 1 year of research experience may also be considered.)
  • Strong background in machine learning, cognitive modeling, or multi-agent systems.
  • Experience in open-source deep learning frameworks (PyTorch, Tensorflow, JAX, etc.)
  • Excellent problem-solving and research skills, with the ability to work independently and collaboratively in an interdisciplinary environment.

 

Bonus Qualifications​

  • ​Familiarity with Large Language Models (LLMs) and Vision-Language(-Action) models for reasoning and communication tasks.
  • Prior experience with multi-agent simulation platforms (e.g., VirtualHome, PsychSim, Habitat, or PARTNR) and experience in dataset synthesis or augmentation for social or collaborative tasks.

 

Years of Work Experience Required  0
Desired Start Date 5/18/2026
Internship Duration 3 Months
Position Keywords ​Agentic AI, Theory of Mind (ToM), Human–AI Collaboration

Alternate Way to Apply

Send an e-mail to careers@honda-ri.com with the following:
- Subject line including the job number(s) you are applying for 
- Recent CV 
- A cover letter highlighting relevant background (Optional)

Please, do not contact our office to inquiry about your application status.