Develop and optimize 3D computer vision components, including triangulation, bundle adjustment, and camera calibration, and integrate them into the GenWeb3D stack.
Design and implement 3D point and sparse patch tracking systems from video data.
Integrate deep learning models into classical vision workflows, including lightweight neural modules that accelerate 3D inference.

Qualifications:

Strong grasp of both classical 3D vision (geometry, structure-from-motion, calibration) and neural computer vision (CNNs, keypoint detection, depth estimation).
Experience with at least one major 3D or perception stack (e.g., ARKit, Meta Reality Labs, Amazon Robotics, Aurora, Waymo, or Boston Dynamics).
Proficient in Python and C++ with experience in deep learning frameworks (PyTorch, TensorFlow).

Collaboration:

Collaborate with the broader engineering and research teams to bring novel 3D perception methods into production-quality systems.
Passionate about building and engineering, not just publishing — you care about robustness, performance, and real-world deployment.

GenWeb

GenWeb is building the next generation of 3D perception systems. As a company, we focus on developing high-performance computer vision pipelines that bridge classical geometry with modern machine learning, aiming to make real-world 3D understanding faster and more accurate. Our team is engineering-focused, valuing robustness and production-quality deployment.

Apply for This Position