About The Role: As an AI Robotics CV Engineer, you are building a robot that must understand the 3D world around it with the same fluency a human does; from precise object localization to spatial reasoning across dynamic environments. This role focuses on the geometric and structural vision systems that underpin everything the robot perceives and acts on.
What You'll Do
- Design and train deep learning models for 3D scene understanding, geometric reasoning, and spatial reconstruction
- Develop vision pipelines for object detection, pose estimation, depth perception, and scene segmentation on-robot
- Build and iterate on occupancy and world-representation networks that operate in real-time on edge hardware
- Create automated data generation pipelines to produce high-quality training data at scale from fleet/robot deployments
- Research and implement state-of-the-art techniques from 3D vision, NeRF, Gaussian splatting, and related fields
- Work closely with hardware and robotics teams to ensure vision systems operate robustly under real-world conditions
What You'll Bring
- MS/PhD or equivalent experience in Computer Vision, Robotics, or Machine Learning
- Strong background in 3D computer vision — stereo, depth estimation, point clouds, SLAM, or structure-from-motion
- Proficiency in PyTorch or JAX; experience training large-scale models end-to-end
- Demonstrated ability to take models from research to production deployment
- Experience with camera calibration, multi-camera systems, and sensor fusion
- Publications or shipped products in geometric vision, embodied AI, or autonomous systems preferred
