Senior Deep Learning Engineer, Cosmo 3D Spatial
Core
Own the data engine and end-to-end training verification for NVIDIA Cosmos 3D spatial reasoning models, ensuring high-quality data and rigorous evaluation for physical AI applications.
Role type
Senior Deep Learning Engineer (Data & Evaluation)
Builds
3D spatial reasoning models for robots, autonomous vehicles, and smart infrastructure
Domain
Physical AI, 3D Computer Vision, Multimodal Learning
Deliverable
production ML models
Required skills
Deep learning system design, 3D computer vision, multi-view geometry, distributed training, large-scale data pipeline engineering, rigorous evaluation methodology, Python, PyTorch/JAX
Preferred skills
PhD in relevant field, publications at top conferences, web-scale video corpus curation, 3D foundation model familiarity, vision-language post-training expertise
Technologies
PyTorch, JAX, Linux, Ray, Spark, Slurm, FSDP, VLMEvalKit
Responsibilities
Source and curate large-scale real-world image/video corpora for vision-language training; Build annotation and auto-labeling pipelines for 3D-grounded supervision; Own data quality end-to-end including deduplication and quality scoring; Verify end-to-end model training pipelines and diagnose regressions; Build and operate 3D spatial evaluation suites and public benchmarks; Partner with research scientists to translate hypotheses into measurable capability
Seniority
Senior, hands-on IC
