Senior Computer Vision Engineer (Localization) at Niantic Spatial in San Francisco, CA

Job description

• Application
• At Niantic Spatial, we’re building the future of physical AI. Powered by a proprietary database of over 30 billion posed images, our groundbreaking mapping technology unlocks a new dimension of interaction and spatial intelligence that helps both humans and machines better understand, represent, navigate, and engage with the real environment. Join us as we build the real-world foundational models for physical AI, so people and machines can work efficiently and safely together.
• As a Senior Computer Vision Engineer on the Applied Algorithms team, you will own significant parts of our Localization technology. As a member of this team, you will be responsible for building the Map representing a living geospatial world model, as well as the centimeter-level accuracy Visual Positioning System (VPS) that allows people and robots to explore the real world. Part of the role will be bridging the gap between R&D and production, turning complex geospatial data into a persistent sense of space and enabling the next generation of spatial AI.

Responsibilities

• System Architecture: Design, develop, and maintain production-grade computer vision systems that power Niantic Spatial’s VPS (Visual Positioning System) and 3D mapping pipelines.
• Algorithmic Innovation: Refine and scale algorithms for our VPS system including, SfM (Structure from Motion) and Feed Forwards models, efficient feature and descriptor extraction to have the best possible accuracy and recall. Moving algorithms from research concepts to hardened production code that can scale for our map of the world.
• Performance Engineering: Optimize complex ML and CV code for maximum efficiency in cloud and mobile environments, ensuring low latency and high-performance execution on GPU/CPU.
• Benchmarking & Evaluation: Create and own the tools and frameworks used to evaluate the quality of our spatial grounding and 3D maps against ground-truth data.
• Technical Leadership: Lead technical design reviews, mentor junior engineers, and serve as a team anchor for resolving complex technical disagreements within the mapping stack.
• Cross-Functional Delivery: Collaborate with Product, Research, and Operations teams to ensure that state-of-the-art computer vision solutions translate into delightful user experiences.

Requirements

• Education: BS, MS, or PhD in Computer Science, Robotics, Computer Vision, or a related technical field (or equivalent professional experience).
• Years of Experience: 5+ years of experience developing and shipping algorithms for image processing, computer vision, or 3D reconstruction.
• Coding Proficiency: Expert-level proficiency in python and/or C++.
• Domain Expertise: Proven track record in designing solutions for Structure from Motion (SfM), VPS or 3D mapping..
• Frameworks & Tools: Deep experience with Deep Learning frameworks (PyTorch or JAX) and version control (Git).
• Required in-office days: 3 days per week
• Experience in planning and leading technical projects from inception to production.
• Significant contributions to open-source CV libraries (OpenCV, COLMAP, etc.).
• Experience with CUDA or shader programming for performance optimization.

Skills

Required

• Version control (Git)
• Computer vision algorithms
• Image processing
• 3D reconstruction
• Structure from Motion (SfM)
• VPS or 3D mapping

Preferred

• CUDA or shader programming

Benefits

• $189K – $255K • bonus + equity + benefits.
• Individual pay within this salary range is determined by work location and additional factors, including assessed job-related skills, experience, and relevant education or training. Your recruiter can answer any questions about new hire total compensation during the hiring process. An overview of benefit offerings for your location can be found on the careers page.