Job Description

In close collaboration with other computer vision team members, this role will involve a healthy mix of classical 3D computer vision and modern deep learning engineering, focused around 3D scanning, 3D reconstruction, SLAM, and scene understanding domains. We don't separate research from development. In this role, you'll be writing production code with focus on quality and value for customers, at the forefront of state-of-the-art computer vision.

What you’ll do:

• Develop new and improve existing computer vision algorithms in the field of house-scale 3D reconstruction (from LiDAR and/or monocular camera), SLAM, and scene understanding.

• Build efficient implementations of algorithms and deliver them to production.

• Remain up to date with the latest advances in computer vision and deep learning technology, and apply them to our products..

• Work closely with the iOS team to improve real-time scanning experience for our scanning app users.

You should have:

• Experience in 3D computer vision R&D such as 3D scanning, RGBD algorithms, visual SLAM, sensor calibration and fusion, multi-view geometry, tracking and pose estimation, etc.

• Ability to quickly evaluate applicability and value of existing state-of-the-art research for our product needs.

• Experience with writing production C++ code.

• Solid math background, especially in 3D math and linear algebra.

• Comfortable working across multiple time zones and cultures.

Nice to have: 

• Deep learning experience in application to 3D computer vision problems.

• Experience of working with large production C++ codebases.