Robotics Perception Engineer

Simbe is building the AI powered operating system for physical retail. Our autonomous robots and multimodal computer vision platform turn complex, constantly changing stores into accurate, actionable intelligence for leading retailers around the world. Simbe combines robotics, computer vision, machine learning, data infrastructure, and customer focused product design to help retailers improve shelf availability, price and promo execution, inventory accuracy, and store team productivity.

Simbe is looking for a Principal or Staff Robotics Perception Engineer to serve as a critical bridge between our Computer Vision and Robot Software teams. This role owns the path from perception models and sensor data to real robotic behavior in production environments. You will help Tally understand dynamic retail stores, improve autonomy, and translate Simbe's computer vision capabilities into more reliable navigation, localization, calibration, and customer facing intelligence.

Why This Role Is High Impact

• You will work on real robots operating in complex, high variance environments, not only offline benchmarks.
• Your work will improve autonomy and customer value across one of the largest real world retail robotics deployments.
• You will connect AI research, production computer vision, robot software, edge inference, and field performance.

Responsibilities

• Own production perception pipelines. Lead the design, evaluation, integration, and optimization of perception systems used by Tally robots, including object detection, semantic and instance segmentation, depth estimation, tracking, sensor fusion, visual localization, and camera calibration.
• Bridge CV and robot autonomy. Partner with Computer Vision, Robotics Software, Product, and Field teams to integrate perception models into the robot codebase and improve navigation, localization, scene understanding, and system reliability.
• Improve real time performance. Profile and optimize neural networks on current and next generation robot hardware using Python, C++, ONNX, TensorRT, CUDA, and embedded GPU workflows.
• Advance robot perception R&D. Explore emerging techniques in visual SLAM, visual odometry, depth estimation, open vocabulary detection and segmentation, multimodal perception, and simulation to identify practical improvements for Simbe's platform.
• Develop system benchmarks. Define and maintain benchmarks that connect model quality to robot outcomes, including latency, compute, uptime, navigation success, localization robustness, false positives, false negatives, and customer facing accuracy.