Fall 2026 Software Co-op

berkshiregrey; Bedford (MA); On-site; USD 24,796 - 38,572; Part time

Job summary

Berkshire Grey is seeking a Software Co-op to join our robotics software team in Bedford, MA, focusing on perception challenges with computer vision and machine learning. You will collaborate with engineers to develop, test, and deploy perception solutions for automated systems.

The role emphasizes evaluation, prototyping, and validation of perception algorithms and sensor systems, with the onsite office in Bedford. July–December 2026 dates are flexible.

Qualifications

• Pursuing a bachelor’s or master’s degree in CS/CE/Robotics or related field.
• Experience programming in Python and/or C++.
• Experience developing in Linux (Ubuntu) and familiarity with ROS.
• Familiarity with robotics and computer vision (classical and/or learning-based).
• Experience with OpenCV or PyTorch.

Responsibilities

• Collaborate with software, hardware, and systems engineers to design, test, and validate perception approaches.
• Support evaluation and testing of perception sensors and benchmarks.
• Develop and test computer vision and ML models (e.g., depth completion, stereo vision, object detection).
• Explore and implement model-based approaches to improve perception pipeline performance.
• Investigate vision-language models (VLMs) for applications like grasp ranking and anomaly detection.
• Utilize simulation tools to evaluate and improve perception models and system performance.
• Develop, debug, and deploy software components for robotic systems in a Linux/ROS environment.
• Document test results and present findings to the engineering team.

Skills

Python

C++

Linux (Ubuntu)

ROS

Robotics & CV familiarity

OO design / data structures

Analytical & problem-solving

Communication skills

Education

Bachelor's/Master's in Computer Engineering, Computer Science, Robotics

Tools

OpenCV

PyTorch

ROS

Job description

The Software Co-op will work as a member of our robotics software team, applying computer vision and machine learning techniques to solve real-world perception challenges in automated systems. This role will collaborate closely with multidisciplinary engineering teams to develop, test, and deploy perception solutions that improve system performance and reliability. Responsibilities will focus on evaluation, prototyping, and validation of perception algorithms and sensor systems.

This role will be onsite at the Bedford, MA office. This position will run from approximately July to December 2026 (dates flexible).

Functions

• Collaborate with software, hardware, and systems engineers to design, test, and validate perception approaches.
• Support evaluation and testing of perception sensors, including cost-down alternatives and performance benchmarking.
• Develop and test computer vision and machine learning models (e.g., depth completion, stereo vision, object detection).
• Explore and implement model-based approaches to improve perception pipeline performance and reduce system bring-up time.
• Investigate the use of vision-language models (VLMs) for applications such as grasp ranking, SKU identification, and anomaly detection.
• Utilize simulation tools to evaluate and improve perception models and system performance.
• Develop, debug, and deploy software components for robotic systems in a Linux/ROS environment.
• Document test results and present findings to the engineering team.

Qualifications

• Actively pursuing a bachelor’s or master’s degree in Computer Engineering, Computer Science, Robotics, or a related field
• Experience programming in Python and/or C++
• Experience developing in Linux (Ubuntu) and familiarity with ROS
• Familiarity with robotics and computer vision (classical and/or learning-based approaches)
• Experience with computer vision or machine learning libraries (e.g., OpenCV, PyTorch)
• Solid understanding of object-oriented design, data structures, and algorithms
• Strong analytical, debugging, and problem-solving skills
• Strong written and verbal communication skills
• Curious, self-motivated, and eager to learn

COOP-2611