Senior Robotics Software Engineer – Object Perception

Agility Robotics is a pioneer. Our robot, Digit, is the first to be sold into workplaces across the globe. Our team is differentiated by its expertise in imagining, engineering, and delivering robots with advanced mobility, dexterity, intelligence, and efficiency -- robots specifically designed to work alongside people, in spaces built for people. Every day, we break through engineering challenges and invent new solutions and capabilities that will one day make robots commonplace and approachable. This work is our passion and our responsibility: our mission is to make businesses more productive and people’s lives more fulfilling. About the Role We are looking for a Senior Robotics Software Engineer to join our Skills team, focusing on object perception for real-world robot manipulation. In this role, you will be close to the real robot, owning perception performance end-to-end on hardware: from sensor inputs, to perception algorithms and models, to how those outputs are consumed by downstream planners to enable robust manipulation in real customer environments. You will work closely with other AI/ML engineers, exploring off-the-shelf models, classical computer vision, geometric reasoning, and learned approaches to build perception systems that are reliable, debuggable, and performant on real robots. In some cases, that will involve training or fine-tuning models; in others, it will mean integrating proven techniques or existing libraries to meet the task requirements efficiently. About the Work Own object perception performance on the robot, from sensors through runtime behavior integration Design, implement, and deploy perception pipelines for manipulation, including object pose estimation, depth reasoning, and scene understanding Work hands-on with real robots to debug, tune, and harden perception systems under real-world variability Collaborate with AI/ML teams to evaluate, integrate, train, and fine-tune perception models Explore and integrate off-the-shelf perception solutions and libraries where they meet performance and reliability needs Combine learned models with classical CV, geometry, and task structure to meet system-level requirements Define interfaces and contracts between perception, behaviors, and task planning to ensure reliable execution Build tools and workflows to evaluate perception performance across datasets, simulation, and hardware About You BS or MS in Robotics or related field 5+ years of professional experience building object perception systems for real robotic platforms Experience working across learned models and classical perception approaches Experience training, fine-tuning, and evaluating perception models Strong experience writing production-quality C++ and Python code Hands-on experience with monocular vision, stereo vision, RGB-D sensing, or depth reconstruction Experience with perception libraries (OpenCV, Open3D, PCL, ROS perception, NumPy, scikit-image) 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 You Might Also Have Solid understanding of spatial reasoning Demonstrated ability to apply object perception model outputs (detection, segmentation, pose estimation) in manipulation pipelines Experience with structured 3D estimation (plane fitting, normal estimation, voxel maps, etc) This is a hybrid role based out of one of our Salem, OR, Pittsburgh, PA, or Fremont, CA offices. The final salary offered to a successful candidate will be dependent on several factors that may include but are not limited to: market location, job-related knowledge, skills, and experience. This range may change based on geographical location and may be modified in the future. Anticipated Salary Range $175,000 $273,000 USD In addition to base pay, our competitive t ... (truncated, view full listing at source)