Senior 3D Spatial AI Engineer

NVIDIA Corporation; Santa Clara (CA); On-site; USD 224,000 - 357,000; Full time

Job summary

NVIDIA is seeking a Senior Deep Learning Engineer to own the data engine and end-to-end training verification for Cosmos spatial reasoning. You will manage sources, annotation, and quality for 3D-grounded training, and you will run large multi-node GPU training to ensure scalable performance.

You will collaborate with Cosmos Lab researchers to translate 3D hypotheses into measurable capabilities, building robust evaluation suites and open benchmarks as part of NVIDIA's Cosmos platform.

Qualifications

• MS or PhD in CS, EE, Robotics, or related field, or equivalent experience.
• 12+ years building deep learning systems in Python with PyTorch or JAX on Linux.
• Deep expertise in 3D computer vision, multi-view geometry, SLAM, depth or pose estimation.
• Hands-on with vision-language models, data and evaluation that improve grounding.
• Experience building large-scale multimodal data pipelines and dataset versioning.
• Experience with distributed training on multi-GPU/multi-node clusters.
• Rigorous evaluation methods and clear communication with researchers.

Responsibilities

• Own the 3D data engine for Cosmos spatial reasoning and curate large-scale corpora.
• Build annotation and auto-labeling pipelines for 3D-grounded supervision at scale.
• Manage data quality end-to-end including deduplication and coverage analysis.
• Verify end-to-end model training and guard reproducibility and regressions.
• Develop and operate 3D and spatial evaluation benchmarks and trace results.

Skills

3D computer vision

Multimodal data pipelines

Distributed training

Python

PyTorch/JAX

Multi-GPU training

Geometry / 3D reconstruction

Visual grounding

Engineering collaboration

Data quality & evaluation

Education

MS or PhD in Computer Science, Electrical/Computer Engineering, Robotics, or related field

Tools

Ray

Spark

Slurm

FSDP

Job description

NVIDIA is seeking a Senior Deep Learning Engineer to own the data engine and end-to-end training verification for Cosmos spatial reasoning. You will manage sources, annotation, and quality for 3D-grounded training, and you will run large multi-node GPU training to ensure scalable performance.

You will collaborate with Cosmos Lab researchers to translate 3D hypotheses into measurable capabilities, building robust evaluation suites and open benchmarks as part of NVIDIA's Cosmos platform.