Matterport – Senior Machine Learning/Computer Vision Engineer – 3D Reconstruction and Semantic Understanding
Core
Research, design, and deploy state-of-the-art machine learning and computer vision systems for 3D reconstruction, semantic understanding, and spatial capture to create high-fidelity digital twins of buildings.
Role type
Senior IC machine learning/computer vision engineer (3D reconstruction and semantic understanding)
Builds
Production-ready 3D assets, digital twins, and spatial capture systems for the built world
Domain
Real estate technology, spatial computing, 3D reconstruction, computer vision
Deliverable
production ML models | product features
Required skills
camera pose estimation, 3D geometry, multi-view reconstruction, deep learning, PyTorch, C++, Python, OpenCV, mesh generation, surface reconstruction, texturing, semantic scene understanding, segmentation, depth estimation, multi-view geometry, appearance modeling, model training, production deployment, data collection strategy, evaluation frameworks, version control (Git), debugging complex systems
Preferred skills
SLAM, structure-from-motion, NeRF, Gaussian Splatting, neural rendering, differentiable rendering, UV mapping, texture generation, GPU performance optimization (CUDA), translating research to production
Technologies
PyTorch, JAX, OpenCV, CUDA, Git
Responsibilities
Research and develop camera pose estimation algorithms; advance mesh generation and texturing pipelines; design intelligent systems for semantic scene understanding; apply deep learning and foundation models to the vision pipeline; collaborate with hardware and product teams to address failure modes; contribute to the full ML lifecycle from data collection to monitoring; prototype novel computer vision techniques; integrate latest advances in neural rendering into production systems; write high-quality Python and C++ code
Seniority
Senior, hands-on IC
