3D Reconstruction Research Scientist/Engineer
Core
Develop and ship 3D reconstruction technology (SfM, 3D Gaussian Splatting, generative models) to create AI-readable representations of the physical world for robots and machines.
Role type
Research Scientist/Engineer (3D Computer Vision & Spatial AI)
Builds
High-fidelity 3D reconstructions, spatial foundation models, and production-ready reconstruction pipelines
Domain
Computer Vision, Machine Learning, Graphics, Robotics, Geospatial
Deliverable
production ML models
Required skills
3D computer vision, neural scene representations, 3D Gaussian Splatting, SfM, SLAM, geometric deep learning, generative models, Python, PyTorch
Preferred skills
C++, CUDA, classical reconstruction tooling (COLMAP), meshing, texturing
Technologies
PyTorch, Python, CUDA, C++, COLMAP
Responsibilities
Conduct original research and publish at top-tier venues (CVPR, ICCV, ECCV, SIGGRAPH); Design and implement novel algorithms for 3D reconstruction and spatial understanding; Transition research ideas into customer-facing products; Build large-scale experimentation and benchmarking frameworks; Collaborate with researchers in CV, ML, graphics, and robotics; Implement and iterate on state-of-the-art methods for academic and production deployment
Seniority
Junior to Senior, research-focused IC
