Senior CV Engineer, 3D Scene Understanding & Embodied AI

Niantic; San Francisco (CA); On-site; USD 230,000 - 255,000; Full time

Job summary

Niantic Spatial seeks a Senior Computer Vision Engineer for its Embodied AI team, focused on 3D scene understanding and semantics. You will work with researchers to turn advances into production-ready capabilities for perception and reconstruction in real environments.

You’ll own end-to-end output quality, adapt methods to real customer data, and help scale systems for secure, reliable deployment across our platform.

Qualifications

• 5+ years of computer vision experience and a bachelor’s degree or equivalent.
• Experience building production CV, ML, graphics, or data-processing systems.
• Hands-on scene understanding: semantic, instance, or panoptic segmentation; open‑vocabulary detection.
• Experience translating 2D perception to 3D: multi‑view fusion, 3D segmentation, scene graph construction.
• Delivered reliable computer‑vision or 3D systems used by other teams or customers.
• Practical command of 3D and spatial data: geometry, camera models, calibration, scale.
• Strong Python and PyTorch; some C++ when needed.
• Experience with cloud infrastructure, GPU workloads, distributed processing, or large‑scale data pipelines.

Responsibilities

• Productionize 3D Scene Understanding Capabilities across varied data and environments.
• Own End-to-End Output Quality for reconstructed environments.
• Bridge Research and Production by testing methods on real data and incorporating constraints.
• Make Systems Robust to the Real World by diagnosing capture quality issues.
• Build Evaluation and Quality Infrastructure with datasets, tests, gates, benchmarks.
• Improve Performance, Throughput, and Cost of GPU/distributed workloads.
• Collaborate with product and engineering teams to align with customer workflows.

Skills

Scene understanding

Python & PyTorch

Open‑vocabulary perception

2D→3D transition

Production-grade systems

Cloud infrastructure

GPU workloads

Education

Bachelor's degree or equivalent

Tools

C++

CUDA

Job description

Niantic Spatial seeks a Senior Computer Vision Engineer for its Embodied AI team, focused on 3D scene understanding and semantics. You will work with researchers to turn advances into production-ready capabilities for perception and reconstruction in real environments.

You’ll own end-to-end output quality, adapt methods to real customer data, and help scale systems for secure, reliable deployment across our platform.