Senior Product Specialist

At Niantic Spatial, we build real-world foundational models for physical AI, so people and machines can work efficiently and safely together.

We give robots, AI agents, and people an accurate, shared understanding of physical spaces - making operations safer and more efficient across some of the world's most demanding environments.

Our reconstruction technology captures environments with geometric accuracy and extreme detail from any standard camera; the Visual Positioning System delivers precise positioning almost anywhere in the world.

We're building for the 80% of economic activity that takes place beyond our screens: robots that lose GPS and drift off course, defense teams with no shared 3D picture of where they operate, facilities where downtime costs millions. Niantic Spatial serves customers in robotics, public sector, and energy and industrial.

As a Senior Product Specialist at Niantic Spatial, you will work at the intersection of 3D reconstruction research, product development, and customer engagements to deliver real-world impact, shaping how we capture, represent, and deliver photorealistic spatial data at a global scale. Driven by our mission to build a living model of the world that people and machines can talk to, you will take a hands-on role in scaling our end-to-end reconstruction capabilities from ground-level capture to real-time rendering. This role has multi-team visibility, directly influencing how our foundation models translate real-world data into photorealistic reconstructions for emerging applications in Physical AI.

• Reconstruction Project Delivery: Partner with Product, Engineering, Research, and Go-to-Market teams to develop and deliver high-fidelity reconstruction Proof of Concepts (PoCs) for lighthouse enterprise customers.

• Working with Customers: Be the point-person for engaging with enterprise customers, interpreting requirements, capturing necessary data, and presenting solutions and results to stakeholders.

• Technical Core: Drive continuous improvements to Niantic Spatial’s 3D reconstruction offering, with a focus on Gaussian splatting, meshing, and hybrid neural representations.

• Sensor & Data Evaluation: Systematically evaluate reconstruction quality across diverse capture modalities, including smartphones, 360 cameras, drones, and LiDAR.

• Quality & Benchmarking: Define, track, and optimize quality metrics for 3D reconstruction outputs; establish rigorous benchmarks that guide internal research and product engineering priorities.

• Landscape Expertise: Stay embedded in the fast-moving 3D reconstruction landscape, keeping up with advancements across the industry to advocate for internal product developments.

• 5-8+ years of industry or applied research experience in 3D reconstruction involving data capture and processing for photogrammetry, NeRFs, and Gaussian Splatting outputs

• Experience with agentic development workflows for building tooling, pipelines, and processes to boost productivity.

• Hands-on expertise and advanced knowledge of 3D reconstruction tooling and workflows such as NerfStudio, COLMAP, RealityScan/RealityCapture, Pix4D, Drone Deploy, Cesium iTwin, and ArcGIS Reality.

• Hands-on expertise with 3D scanning. Advanced capture methods, scan project planning, drone flight path planning, execution, and data organization and management.

• Strong understanding of the full reconstruction stack: capture methods and devices, SfM and SLAM, dense reconstruction, and an understanding of optimizing reconstructions for real-time rendering.

• Experience with multi-sensor data (RGB, depth, LiDAR, IMU) and the tradeoffs between capture modalities.

• Presentation and technical writing skills to present to both deeply technical and non-technical audiences

• Required at least 3 days of in-office work in San Francisco.

• Bachelor’s Degree in Computer Science, Robotics, Computer Vision, or a related technical field

• Experience with program management tools like JIRA, Linear, Confluence

• Familiarity with compression, streaming, and LOD techniques for large-scale 3D assets

• Publications or open-source contributions in 3D vision, computer graphics, or neural rendering