Spatial Computing Engineer: Real-Time 3D Vision & AI

Schemata; San Francisco (CA); On-site; USD 185,000 - 240,000; Full time

Benefits offered by this job

Equity stake

Health coverage

Competitive salary

401(k) plan

Paid parental leave

High visibility and impact

Job summary

Schemata is seeking a highly-specialized Spatial Computing Engineer in San Francisco to design and optimize 3D and spatial intelligence pipelines that turn raw capture data into actionable world models for next-generation simulation and training.

You will work end-to-end from research prototypes to production inference, collaborating across graphics, software, and product teams to deliver real-time, robust spatial solutions, with opportunities to publish and present at top conferences.

Qualifications

• PhD with 3DGS-related work or 4+ years industry experience in 3D computer vision, robotics, graphics, segmentation, or VLM/MLLM model training and application.
• Strong programming skills with applied deep‑learning frameworks.
• Hands‑on experience with 3D data types: point clouds, meshes, 3DGS representations.
• Solid grounding in linear algebra, geometry and numerical optimization.
• Proven ability to translate research into production code and services.
• Experience profiling GPU workloads and scaling training or real‑time inference.

Responsibilities

• Research, prototype, and fine-tune state‑of‑the‑art methods for 3D reconstruction, segmentation, and feature embedding (e.g., 3DGS, DINO, RoPE, FlowR models).
• Design data pipelines that convert heterogeneous inputs (CAD, LiDAR, mesh, simulation output) into unified, queryable scene‑graphs.
• Integrate multimodal foundation models (LLMs, VLMs) with spatial representations to power real‑time diagnostics and autonomous evaluation in training scenarios.
• Implement GPU‑accelerated training/inference, synthetic‑data generation, and large‑scale evaluation workflows in cloud and edge environments.
• Collaborate with graphics and product engineers to ship mission‑critical features that blend neural perception with real‑time rendering.
• Publish internally, attend CVPR / NeurIPS / SIGGRAPH, and translate the latest research into production capabilities.

Skills

PhD in 3DGS

3D computer vision

Robotics

Graphics

VLM/MLLM training

Deep learning

Education

PhD in a relevant field

Tools

CUDA

Python

TensorFlow/PyTorch

Job description

Schemata is seeking a highly-specialized Spatial Computing Engineer in San Francisco to design and optimize 3D and spatial intelligence pipelines that turn raw capture data into actionable world models for next-generation simulation and training.

You will work end-to-end from research prototypes to production inference, collaborating across graphics, software, and product teams to deliver real-time, robust spatial solutions, with opportunities to publish and present at top conferences.