SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Berkshire Grey is a leader in AI and robotics, automating complex warehouse and logistics operations including picking, packing, sorting, and trailer unloading for e-commerce and retail. As a Senior Software Engineer in the Research & Advanced Development group, you will develop novel approaches to solving challenging manipulation problems for real-world robotics systems, enabling them to understand and interact with their environment in unprecedented ways.
You will work across the full robotics stack, from low-level robot drivers to multi-robot orchestration systems running in Berkshire Grey's in-house proving ground. Key responsibilities include training and adapting machine learning architectures for robot learning and visual scene understanding, developing and improving the robot policy learning data collection application, and determining which demonstrations, robot logs, sensor data, and annotations are suitable for fine-tuning robot foundation models when applied to new tasks, embodiments, or environments.
You will stay current with advancements in robotics and related fields, evaluate their applicability to company challenges, and collaborate with external research partners. The role includes mentoring junior engineers and interns, rapidly prototyping solutions with demos for stakeholders, and owning solutions from prototype through production-ready deployment.
Minimum qualifications include a Master's degree in Robotics, Machine Learning, Computer Vision, Computer Science, or related field; software development experience in robotics or automation; strong expertise in Python and C++; experience with Git, CI/CD, ROS/ROS2; and demonstrated ability to design and implement software and algorithms for real robots. You must work effectively in fast-paced environments, communicate technical priorities clearly, rapidly prototype solutions, and explain complex ideas to both technical and non-technical stakeholders.
Preferred qualifications include 3+ years of relevant robotics or automation software engineering experience, expertise in computer vision for robotic manipulation (grasp estimation, object detection, anomaly detection), motion planning and optimization, imitation learning, deep learning frameworks (PyTorch), robot simulators (Isaac Sim, MuJoCo), robot learning frameworks, robotic sensors, model-based and data-driven approaches, Docker/cloud computing, parallel/distributed systems, production robot deployment, and agentic coding tools. Occasional travel up to 10% required.
