3D Computer Vision Research Engineer, Geospatial Visual Localization
Core
Develop novel visual localization algorithms and train large-scale multi-modal models to enable robots and machines to accurately localize themselves anywhere on the planet using 3D vision and deep learning.
Role type
Senior Research Engineer (Geospatial Visual Localization)
Builds
Production-ready visual localization systems and benchmarks for robots and machines operating in the physical world
Domain
Geospatial AI, 3D Computer Vision, Robotics
Deliverable
production ML models
Required skills
3D computer vision, deep learning, multi-view geometry, transformer models for 3D vision, Python, PyTorch, NumPy, SciPy, software engineering, algorithm deployment
Preferred skills
PhD in computer vision/robotics/ML, publication record at top conferences, experience with aerial/satellite imagery, GNSS/GPS, IMUs, LiDAR, geodesy, mapping systems
Technologies
PyTorch, NumPy, SciPy, 3D vision foundation models, multi-modal attention mechanisms
Responsibilities
Design and implement novel visual localization algorithms; train large models on multi-modal data at scale; develop robust solutions for GPS-denied environments and varying capture quality; collaborate with researchers and engineering teams to deploy research at production scale; communicate results through technical writing and publications
