Senior Software Engineer, Agility Robotics

• This role is ideal for someone who wants to be close to the real system, enjoys solving end-to-end integration challenges, and thrives on making perception work reliably in production
• BS or MS in Robotics or related field
• 5+ years of professional experience integrating perception models in real-world robotic applications
• Strong experience writing production-quality Python code
• Hands-on experience working with point clouds, RGBD cameras, and depth sensing in applied robotics contexts
• Experience with perception libraries (OpenCV, Open3D, PCL, ROS perception, NumPy, scikit-image)
• You’re pragmatic: you know when to use task structure and prior knowledge to get the job done reliably
• You love solving hard perception problems, but your real passion is seeing those solutions work on a robot in the real world and having an impact
• You’re excited to work on hardware
• You enjoy debugging, testing, and iterating quickly in simulation and on the real robot

Desirable

• We are looking for a Senior Software Engineer to join our Skills team, with a focus on integrating perception models into real-world robotic applications
• In this role, you will be responsible for perception integration efforts to build robust, structured world representations that can be manipulated and reasoned about in robot workflows
• You will own the performance of the perception stack on the robot, working hands-on with real hardware and validating your work in production workflows
• You’ll collaborate closely with AIML teams and integrate both in-house and off-the-shelf methods to meet customer and task-specific requirements
• Develop robust perception algorithms and applications that leverage existing perception models and structure in the robot task (e.g., stacks, pallets, conveyors, flowracks) to improve robustness and simplify estimation
• Define and maintain reusable APIs that allow the task planner and skills layer to interact with a library of perception skills
• Integrate perception functions with other robot skills like manipulation behaviors using state machines, behavior trees, and task planners
• Manage datasets and run experiments to validate perception skills across diverse task conditions and variations
• Create modular interfaces for field engineers to tune perception skills within your validated specs without requiring software redeployment
• Contribute to defining our overall perception strategy within the skills layer, including library structure, abstraction layers, and tooling