AI Engineer, Computer Vision

Niantic Spatial is building a centimeter-level Visual Positioning System, and the AI Engineer, Computer Vision role focuses on improving deep learning visual localization models from end to end. You will develop and iterate on models, connect field learnings back into training, and support the path from checkpoints to production serving. The work spans PyTorch training and evaluation, high-throughput dataset pipelines, and localization performance measurement under real-world conditions.

What you will do

• Implement, train, and evaluate visual localization models in PyTorch, including local feature extraction and matching, image retrieval, pose regression, and scene-coordinate regression.
• Iterate on architectures, loss functions, and training loops to improve accuracy on real production captures.
• Help with high-throughput ingestion and preprocessing pipelines that convert raw imagery, poses, and 3D point clouds into training-ready datasets, including curation, sampling, and augmentation.
• Run training jobs across GPU environments on a 30-billion-image corpus, partnering with senior engineers and ML Infrastructure to keep experiments reproducible and results interpretable.
• Build and maintain an evaluation suite for localization, including pose accuracy, recall across lighting and seasonal changes, and latency under production load.
• Work with senior engineers to move models from checkpoint toward serving through export, inference optimization, and monitoring once live.
• Own models end-to-end, including learning how field failures translate into training data and benchmark improvements with the Research team and Real-World Test Lab.

Requirements

• Experience training deep learning models for computer vision tasks in PyTorch through coursework, research, internships, or early industry work.
• Foundational understanding of at least one: visual localization, learned feature matching, image retrieval, pose estimation, SLAM, or structure-from-motion, plus underlying 3D geometry (camera models, epipolar geometry, pose parameterizations).
• Comfort working with large image or 3D datasets, including preprocessing, curation, and understanding how data quality affects model behavior.
• Familiarity with GPU-based training and the ability to read and reason about training runs to identify bottlenecks in data loading, compute, or model architecture.
• Strong Python skills and a habit of writing code others can rerun and build on.
• A bachelor’s or master’s degree in computer science, computer vision, robotics, or a related field, or equivalent practical experience.

Technologies

• PyTorch
• Python
• GPU

Compensation & benefits

• Expected salary range: $165,600 - $221,000 per year
• Annual bonus
• Equity
• Comprehensive benefits package including medical, dental, and vision coverage
• 401(k)
• More

Nice to have

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

Location & work model

This role is based in the San Francisco, CA office with a hybrid schedule requiring a minimum of 3 days per week in office.

Inclusive application

Niantic Spatial encourages candidates to apply even if they do not match every qualification listed, as strong candidates may not tick every box.