Robot Learning R&D Co-op — Spring 2027, Onsite Bedford

Berkshire Grey, Inc.; Bedford (MA); On-site; USD 28,000 - 42,000; Part time

Job summary

Berkshire Grey, Inc. invites a Spring 2027 Robot Learning R&D Co-op to work onsite in Bedford, MA, January–June 2027.

You will help develop and integrate new manipulation approaches for real-world robotics systems, spanning robot drivers to the in-house multi-robot orchestrator and proving ground. The role targets advanced-degree students with strong Python and ROS experience, plus ML methods such as imitation or reinforcement learning, applied to physical-world robotics.

Qualifications

• Pursuing an advanced degree (MS or PhD) in Robotics, ML, CS, or related field.
• Strong Python software development; C++ is a plus.
• Experience with ML methods such as imitation or reinforcement learning and vision transformers.
• Experience applying ML to physical-world robotic systems.
• Robotics software development using Linux (Ubuntu) and ROS.

Responsibilities

• Build software infrastructure to integrate, deploy, and evaluate learning-based manipulation capabilities.
• Contribute to training and adaptation of ML architectures for robot learning and scene understanding.
• Contribute to robot learning data collection application development and improvements.
• Build tools to process, validate, visualize, and evaluate data from on-robot policy execution (logs, sensors, annotations).
• Integrate novel grippers and hardware into perception, planning, and robot-control systems.
• Leverage production data and internal systems to implement and evaluate solutions.

Skills

Python

C++

Imitation learning

Reinforcement learning

Vision transformers

Robotics ML

Education

MS/PhD candidate in Robotics/ML/CS

Tools

ROS

Linux

Job description

Berkshire Grey, Inc. invites a Spring 2027 Robot Learning R&D Co-op to work onsite in Bedford, MA, January–June 2027.

You will help develop and integrate new manipulation approaches for real-world robotics systems, spanning robot drivers to the in-house multi-robot orchestrator and proving ground. The role targets advanced-degree students with strong Python and ROS experience, plus ML methods such as imitation or reinforcement learning, applied to physical-world robotics.