Senior Computer Vision Engineer, Embodied AI

Niantic Spatial•San Francisco, CA

2d•Hybrid

About The Position

Niantic Spatial is building the future of physical AI with proprietary mapping technology that unlocks a new dimension of interaction and spatial intelligence. Our reconstruction technology captures environments with geometric accuracy and extreme detail, and our Visual Positioning System delivers precise positioning globally. We serve customers in robotics, public sector, and energy/industrial markets. This role focuses on bridging the sim-to-real gap for visual-spatial understanding in embodied AI, leveraging decades of experience in encoding the world. The Senior Computer Vision Engineer will turn advances in 3D reconstruction and spatial perception into reliable capabilities for embodied AI teams in production, focusing on practical aspects like inputs, edge cases, operational constraints, and downstream requirements. This is a hands-on engineering role where the individual will build, ship, and own outcomes, shaping both technology and product while staying close to the code and customer problems.

Requirements

Significant experience building and operating production computer-vision, machine-learning, graphics, or data-processing systems.
Hands-on experience with 3D reconstruction, photogrammetry, structure from motion, SLAM, mapping, scene understanding, computer graphics, or a related field.
Experience delivering reliable computer-vision or 3D systems used by other teams or customers.
Practical command of 3D and spatial data, including geometry, camera models, coordinate frames, calibration, and metric scale.
Strong Python skills, with the ability to work in C++ or other performance-oriented environments when needed.
Experience with cloud infrastructure, GPU workloads, distributed processing, or large-scale data pipelines.
A bachelor's degree in a relevant field, or equivalent experience.

Nice To Haves

Experience with Gaussian splatting, neural rendering, meshing, or related reconstruction methods.
Built evaluation or benchmarking infrastructure for perception or reconstruction systems.
Experience with robotics, simulation, or embodied AI applications.
Experience optimizing large-scale GPU workloads for cost, throughput, or latency.

Responsibilities

Productionize Computer-Vision and 3D Reconstruction Capabilities: Turn research ideas and prototypes into reliable pipeline components and services that can operate across varied customer data and environments.
Own End-to-End Output Quality: Ensure that reconstructed environments and spatial artifacts are geometrically coherent, correctly scaled and aligned, and usable by downstream simulation and embodied AI systems.
Make Systems Robust to the Real World: Diagnose failures caused by capture quality, scene complexity, scale, calibration, coordinate conventions, and other assumptions that prototypes often leave implicit.
Improve Performance, Throughput, and Cost: Profile and optimize GPU and distributed workloads, reduce unnecessary reruns, and help establish the economics of running reconstruction at scale.
Build Evaluation and Quality Infrastructure: Create datasets, regression tests, quality gates, and benchmarking tools that help the team measure whether changes improve the system.
Shape the Next Generation of Capabilities: Bring production and customer constraints into research planning, helping prioritize the advances that matter most for embodied AI.
Work Across the Product Boundary: Partner with Embodied AI product and engineering teams to understand customer requirements and deliver capabilities that fit real training, evaluation, and deployment workflows.
Define the Interfaces to Embodied AI Workflows: Help evolve the representations, tooling, and operational systems that allow reconstructed environments to move reliably into customer applications.