Quick overview
Job description
About the Role
Join an industrial inventory measurement team building a computer vision system that turns stereo camera data into accurate, auditable pile measurements. You will own the core perception pipeline and help bring research methods into production-grade measurement.
What You'll Do
Develop stereo reconstruction, depth estimation, segmentation, and point-cloud processing workflows.
Improve 3D reconstruction quality, volume accuracy, calibration assumptions, and measurement confidence metrics.
Build and maintain ground-plane estimation, point-cloud cleanup, stitching, and volume calculation components.
Integrate and evaluate vision models, including stereo and YOLO-style approaches.
Analyze failure modes and debug perception systems using real-world sensor data.
What We're Looking For
Four or more years of experience in computer vision or perception machine learning engineering.
Strong Python and PyTorch skills, with experience using OpenCV and Open3D.
Hands-on experience with stereo vision, depth estimation, camera calibration, point clouds, 3D geometry, or reconstruction.
Experience shipping computer vision systems that use real-world sensor data.
Background in stereo perception and volume reconstruction through work or research.
Experience with CUDA, robotics perception, SLAM, photogrammetry, industrial inspection, or geospatial and volumetric measurement is beneficial.
Compensation & Benefits
Base salary: $180,000 to $245,000 annually. Visa sponsorship is available.
Location
On-site in San Francisco, California, United States.