AI Engineer, Computer Vision
Niantic Spatial11 hours ago
Base Salary
$166k - $221k/yr
Responsibilities
- Implement, train, and evaluate visual localization models in PyTorch, including feature extraction, feature matching, image retrieval, pose regression, and scene-coordinate regression.
- Build high-throughput ingestion and preprocessing pipelines for imagery, poses, and 3D point clouds, including data curation, sampling, and augmentation.
- Run reproducible training experiments across GPU environments on a 30-billion-image corpus.
- Develop and maintain localization benchmarks covering pose accuracy, recall across environmental changes, and production latency.
- Support model export, inference optimization, monitoring, and deployment into production services.
- Collaborate with research, engineering, ML Infrastructure, and the Real-World Test Lab to connect field failures with training and benchmark improvements.
Requirements
- Experience training deep learning models for computer vision in PyTorch through coursework, research, internships, or early industry work.
- Foundational knowledge of visual localization, learned feature matching, image retrieval, pose estimation, SLAM, or structure-from-motion, including relevant 3D geometry.
- Experience working with large image or 3D datasets and understanding how data quality affects model behavior.
- Familiarity with GPU-based training and interpreting training runs to identify data-loading, compute, or model-architecture bottlenecks.
- Strong Python skills and the ability to write reproducible, maintainable code.
- Bachelor’s or master’s degree in computer science, computer vision, robotics, or a related field, or equivalent practical experience.
- Preferred qualifications include publications or contributions in learned localization or related vision research, large-scale self-supervised or multi-view pretraining, vision foundation model fine-tuning, petabyte-scale spatial data, robotics or augmented reality experience, edge or mobile inference optimization, and production computer vision internship experience.
Benefits
- Medical, dental, and vision coverage.
- 401(k) and additional benefits.
- Hybrid work model based in the San Francisco, California office, with a minimum of three days per week in the office.
Categories
About Niantic Spatial
Niantic Spatial builds computer‑vision and spatial computing software that maps real‑world environments and localizes people and machines within them. The company offers a platform with APIs and SDKs for creating AI‑native digital twins, visual positioning, and spatial question‑answering, serving robotics, AR/VR, defense, and industrial operations. Privately held, it sells its technology to enterprises that need precise, geometrically accurate understanding of physical spaces beyond GPS or screen‑bound systems.