Research Scientist, Embodied AI & Spatial Intelligence

Javelin Venture Partners; Greater London; Hybrid; GBP 104,000 - 203,000; Full time

Job summary

Niantic Spatial in London is seeking a Research Scientist to advance Spatial AI and contribute to the Large Geospatial Model within the Embodied AI group. You will invent algorithms at the intersection of computer vision, ML, graphics, and robotics, with a focus on real-to-sim pipelines and production-ready prototyping.

The role requires a PhD or equivalent research background with a strong publication record, experience in sim-to-real transfer, and proficiency in Python and PyTorch.

Qualifications

• PhD or equivalent research experience with a strong publication record.
• Substantial expertise in robotics or embodied AI, incl. sim-to-real transfer and domain randomization.
• 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).
• Experience with 3D semantics, open-vocabulary scene understanding, or scene graphs.
• Proficiency in Python and PyTorch.
• GPU/graphics programming with CUDA, shaders, or differentiable rendering.
• Track record of contributing to open-source research codebases.

Responsibilities

• Develop new spatial representations for embodied AI to learn from, train, and act on.
• Advance real-to-sim pipelines connecting real-world observations to 3D reconstruction and robotic policy learning.
• Solve spatial intelligence problems related to 3D scene representation and model efficiency.
• Take new capabilities through to evidence: assess policy impact and reconstruction changes.
• Build prototypes, eventually running at production scale on customer data.
• Collaborate with researchers, engineers, and product teams to bring models to production.
• Publish and present your work.

Skills

Python

Deep Learning

3D Vision

Research Skills

Collaboration

Education

PhD in computer vision/robotics/graphics/ML or related field

Tools

PyTorch

CUDA

Shaders

Differentiable rendering

Job description

Niantic Spatial in London is seeking a Research Scientist to advance Spatial AI and contribute to the Large Geospatial Model within the Embodied AI group. You will invent algorithms at the intersection of computer vision, ML, graphics, and robotics, with a focus on real-to-sim pipelines and production-ready prototyping.

The role requires a PhD or equivalent research background with a strong publication record, experience in sim-to-real transfer, and proficiency in Python and PyTorch.