Embodied AI Research Scientist – Spatial Vision (London)

Niantic Spatial, Inc.; Greater London; Hybrid; GBP 104,000 - 203,000; Full time

Benefits offered by this job

Annual bonus

Equity

Benefits package

Job summary

Niantic Spatial's London-based R&D team is seeking a Research Scientist to push the frontiers of Spatial AI, advancing our Large Geospatial Model as part of the Embodied AI group. You will invent and develop algorithms spanning computer vision, ML, graphics, and robotics.

We value scientists who learn, publish, teach, and deploy each invention. This role is based in London with three days per week in office.

Qualifications

• PhD in computer vision, robotics, graphics, ML or related field with a publication record.
• Experience in robotics or embodied AI, including sim-to-real transfer, domain randomization, and policy evaluation.
• Experience with 3D scene representations: Gaussian splatting, neural fields, meshes, point clouds, or implicit representations.
• Applied generative or diffusion models to 3D data (scene completion, inpainting, relighting, or novel view synthesis).
• Worked on 3D semantics, open-vocabulary scene understanding, or scene graphs.
• Proficiency in Python and PyTorch.
• GPU/graphics programming (CUDA kernels, shaders, differentiable rendering).
• Track record of contributing to open-source research codebases.

Responsibilities

• Develop new spatial representations for embodied AI to learn, train, and act on.
• Advance real-to-sim pipelines connecting real-world observations, 3D reconstruction, simulation, and robotic policy learning.
• Solve emerging spatial problems related to spatial intelligence and model efficiency.
• Take capabilities from development to evidence: show impact on policy and reconstruction.
• Build prototypes, initially for internal use, with some production-scale deployments.
• Collaborate with researchers, engineers, and product partners to bring models to production.
• Publish and present your work.

Skills

Python

PyTorch

3D representations

Gaussian splatting

Neural fields

Diffusion models

Open-source contributions

Education

PhD in computer vision/robotics/ML or related field

Tools

CUDA

Shaders

Git

Job description

Niantic Spatial's London-based R&D team is seeking a Research Scientist to push the frontiers of Spatial AI, advancing our Large Geospatial Model as part of the Embodied AI group. You will invent and develop algorithms spanning computer vision, ML, graphics, and robotics.

We value scientists who learn, publish, teach, and deploy each invention. This role is based in London with three days per week in office.