Research Engineer, SLAM & Multi-View Geometry

OpenAI • San Francisco, CA

1d • Hybrid

About The Position

As a SLAM / Multi-View Geometry Engineer on the Robotics team, you will develop systems that enable robots to perceive, track, and reconstruct the world in 3D from multi-camera and multimodal sensor data. You will work on real-time and offline SLAM pipelines used during teleoperation and robot data collection, as well as scalable systems for reconstructing and tracking 3D structure from large datasets.
We’re looking for people who combine strong fundamentals in computer vision with practical experience building robust perception systems. The ideal candidate is comfortable working across classical geometry-based approaches and modern machine learning methods, and enjoys working closely with AI researchers and engineers.
This role is based in San Francisco, CA. We use a hybrid work model of 4 days in the office per week and offer relocation assistance to new employees.

Requirements

• Have industry experience applying SLAM or visual-inertial odometry systems, such as in robotics, self-driving vehicles, AR/VR headsets, or other real-world perception systems.
• Have a deep understanding of multi-view geometry, camera calibration, bundle adjustment, feature tracking, and sources of error in real-world SLAM systems.
• Have experience with large-scale data processing pipelines or are excited to learn and work with infrastructure that supports large datasets.
• Enjoy working in fast-moving environments and collaborating closely with engineers and researchers to ship systems quickly.

Responsibilities

• Develop and deploy online SLAM systems used during robotic data collection with multi-camera sensor stacks and teleoperation platforms.
• Build systems for large-scale 3D reconstruction and point tracking across massive datasets, enabling new approaches to world modeling and perception.
• Work with research and engineering teams to scale multi-view geometry pipelines to large datasets.
• Improve the accuracy, robustness, and scalability of perception systems used in robotics data collection and training pipelines.
• Collaborate across robotics, perception, and ML teams to integrate geometry-based methods with learned models.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Education Level: No Education Listed