Multimodal 3D Vision & LLM Researcher

Javelin Venture Partners; Greater London; On-site; GBP 120,000 - 180,000; Full time

Job summary

Niantic Spatial in London is seeking a Computer Vision Researcher with a strong background in Large Language Models to bridge 3D vision and language. You will work on aligning spatial geometry with language to enable context-aware navigation and open-ended reasoning about real-world environments.

The role involves leading semantic grounding research, developing continuous semantic capabilities for evolving 3D maps, and mentoring the London R&D hub team.

Qualifications

• PhD (or equivalent) in Computer Vision, ML or Robotics with a focus on Multimodal/Semantic understanding.
• 4+ years of ML research experience with a track record of shipping models bridging 3D vision and language.
• Expert knowledge of 3D geometry (SfM, SLAM, VPS) and Transformer-based architectures (VLMs).
• First-author publications at top-tier venues (CVPR, NeurIPS, ICLR) focusing on VLMs, scene understanding or semantic segmentation.

Responsibilities

• Architect Semantic Grounding: lead research into cross-modal grounding connecting 3D spatial features with language embeddings.
• Scale Understand Capabilities: develop algorithms for continuous semantics for evolving 3D maps.
• Agentic Frameworks: build the spatial brain for Embodied AI enabling robots and drones to perform mission-level reasoning.
• Multimodal Benchmarking: define standards for spatial common sense in LLMs; create evaluations for 3D scene interpretation.

Skills

3D geometry

Transformer-based architectures

Production-quality code (PyTorch/JAX)

Education

PhD in Computer Vision / ML / Robotics

Tools

PyTorch

JAX

ROS

Job description

Niantic Spatial in London is seeking a Computer Vision Researcher with a strong background in Large Language Models to bridge 3D vision and language. You will work on aligning spatial geometry with language to enable context-aware navigation and open-ended reasoning about real-world environments.

The role involves leading semantic grounding research, developing continuous semantic capabilities for evolving 3D maps, and mentoring the London R&D hub team.