Research Intern: Online Learning for Dexterous Manipulation - Honda Research Institute USA

Research Intern: Online Learning for Dexterous Manipulation

Your application is being processed

Research Intern: Online Learning for Dexterous Manipulation

Job Number: P26INT-27
Honda Research Institute USA (HRI-US) is seeking a self-motivated Ph.D. student to join our Intelligence Robotics Research Division as a summer intern. Our research focuses on online learning for dexterous manipulation in real-world, aiming to advance adaptive and data-efficient robotic learning. The successful candidate will work on topics such as increasing efficiency in real-world reinforcement learning frameworks or online control methods.
San Jose, CA

 

Key Responsibilities

 
  • Develop novel algorithms for online learning of dexterous manipulation skills in the real world focusing on aspects such as learning efficiency, model adaptation, reliability, and robustness.
  • Design and perform systematic experiments on real robotic hardware to validate and compare developed methods.
  • Contribute to the software stack by developing efficient and reliable components.
  • Document findings and contribute to internal research reports.
  • Contribute to the portfolio of patents and publish research results when applicable at top-tier conferences and journals in robotics and machine learning.

 

Minimum Qualifications

 
  • Currently enrolled in a Ph.D. program in Robotics, Computer Science, Mechanical Engineering, or a related field, with significant experience in the field.
  • Experience with physics simulators (MuJoCo preferred).
  • Experience in reinforcement learning (RL) and/or model predictive control (MPC).
  • Hands-on experience implementing algorithms and conducting experiments with real robotic hardware.
  • Proficiency in C++ and Python and agentic coding.

 

Bonus Qualifications​

  • Hands-on experience with multi-fingered robotic hands.
  • Experience in real-world reinforcement learning or fine-tuning.
  • Experience with sampling-based MPC in the real world.
  • Experience with GPU-accelerated simulation and domain randomization.

 

Years of Work Experience Required   0
Desired Start Date  May 2027
Internship Duration  3 Months
Position Keywords  

​ Dexterous manipulation, Online Learning, Reinforcement Learning, Model Predictive Control

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.