AI Engineer, Computer Vision
Core
Train and evaluate deep learning models for visual localization, including feature extraction, matching, pose regression, and image retrieval to enable precise positioning in the real world.
Role type
Senior IC computer vision engineer (localization)
Builds
Visual Positioning System (VPS) and geospatial world models for robotics, public sector, and industrial markets
Domain
Computer Vision / Geospatial AI / Robotics
Deliverable
production ML models
Required skills
PyTorch, deep learning model training, visual localization, learned feature matching, image retrieval, pose estimation, SLAM, 3D geometry, Python, GPU-based training, large dataset preprocessing
Preferred skills
published research in localization, self-supervised pretraining, inference optimization (TensorRT/ONNX), petabyte-scale spatial data, robotics, augmented reality
Technologies
PyTorch, Python, GPU environments, TensorRT, ONNX, CoreML
Responsibilities
Train and evaluate visual localization models; build data ingestion and preprocessing pipelines; run large-scale training experiments; maintain evaluation benchmarks; support production deployment and inference optimization; bridge research findings with product requirements
Seniority
Senior, hands-on IC
