3D Computer Vision Engineer

3D Computer Vision Engineer

We are looking for a 3D Computer Vision Engineer to develop the geometric backbone of our scanning technology. You will work on multi-camera calibration, feature-based reconstruction, and geometric optimization—combining classical techniques with modern learned approaches to achieve sub-millimeter accuracy at production scale.

This role is for engineers who love the elegance of projective geometry but aren’t afraid to integrate neural components where they outperform classical methods. If you want to build 3D systems that power tens of thousands of assets for the world’s leading brands, this is for you.

Key Responsibilities

Development of 3D Pipeline Components

• Develop and optimize geometric reconstruction modules: multi-camera calibration, feature extraction and matching, structure-from-motion.
• Implement and tune optimization techniques (bundle adjustment, surface fitting, mesh refinement) for high-precision geometry.
• Integrate learned components—feature descriptors, depth priors, outlier rejection—into classical pipelines.
• Develop capture and calibration strategies that support downstream appearance and material reconstruction.
• Build robust point cloud and mesh processing workflows for cleaning, filtering, and topology repair.
• Write performant, maintainable code that scales across our production infrastructure.

Collaboration

• Work with ML engineers, rendering specialists, and system developers to bridge classical and neural approaches.
• Engage with enterprise clients to understand accuracy and performance requirements.
• Contribute to architectural decisions for scalable 3D processing.

Qualifications

Required:

• Strong foundation in 3D vision and geometric computing, gained through a PhD or Master’s degree combined with 3+ years of industry experience.
• Strong background in multi-view geometry: camera models, epipolar geometry, bundle adjustment.
• Experience with SfM/SLAM pipelines or structured light systems.
• Proficiency in C++ and Python; comfort with performance-critical code.
• Familiarity with point cloud and mesh processing (Open3D, PCL, CGAL, or similar).

Nice to Have:

• Experience with learned feature descriptors or neural depth estimation.
• Familiarity with differentiable optimization.
• Understanding of appearance capture, light calibration, or inverse rendering fundamentals.
• Background in deformable geometry or non-rigid reconstruction.
• Experience with camera/projector calibration or photogrammetry pipelines.
• Familiarity with CUDA or GPU-accelerated geometry processing.
• Contributions to open-source 3D vision projects.