AI Engineer, Computer Vision

Sourced from feeny.ai

Niantic Spatial is hiring an AI Engineer for Computer Vision to help build the deep learning models behind its centimeter-level Visual Positioning System. You will train and evaluate visual localization models in PyTorch, build data pipelines over a 30-billion-image corpus, and help take models from checkpoint to production serving. The role sits at the junction of research and production, working closely with researchers, ML infrastructure engineers, and the Real-World Test Lab.

• Implement, train, and evaluate visual localization models in PyTorch, including feature extraction and matching, image retrieval, pose regression, and scene-coordinate regression
• Contribute to high-throughput ingestion and preprocessing pipelines for imagery, poses, and 3D point clouds
• Run training jobs across GPU environments on a 30-billion-image corpus and keep experiments reproducible
• Build and maintain the localization evaluation suite covering pose accuracy, recall, and latency
• Support production deployment including export, inference optimization, and monitoring
• Translate field failures into training data and benchmark improvements with Research and the Real-World Test Lab

• Experience training deep learning models for computer vision tasks in PyTorch
• Foundational understanding of visual localization, learned feature matching, image retrieval, pose estimation, SLAM, or structure-from-motion, plus underlying 3D geometry
• Comfort working with large image or 3D datasets, including preprocessing and curation
• Familiarity with GPU-based training and ability to reason about training runs
• Strong Python skills and a habit of writing reproducible code
• Bachelor's or master's degree in computer science, computer vision, robotics, or related field, or equivalent practical experience

• Published work on learned localization, visual place recognition, scene-coordinate regression, or vision foundation models (CVPR, ECCV, ICCV, NeurIPS, 3DV)
• Experience with large-scale self-supervised or multi-view pretraining, or fine-tuning vision foundation models for geometric tasks
• Inference optimization for edge or mobile hardware (TensorRT, ONNX, CoreML)
• Experience with petabyte-scale spatial data such as point clouds, meshes, or geospatial indexing
• Background in robotics, augmented reality, 3D reconstruction, digital twins, or autonomous systems
• Prior industry or research internship experience contributing to a production CV system