Senior Vision Language Model Engineer

What you'll do

• Design and implement agentic data workflows to automate data discovery and labeling.
• Build high-quality multimodal datasets tailored for end-to-end physical AI problems.
• Explore and productize new data sources including simulation and synthetic data.
• Partner with researchers to develop prototypes of VLMs and VLAs for video understanding.
• Contribute to NVIDIA Cosmos Dataset Search and other core platforms and products.

What we're looking for

• Strong background in modern deep learning, including transformer-based architectures.
• Extensive experience training and deploying deep learning models on real-world datasets.
• Expertise in Python and at least one deep learning framework.
• Current with latest research in image and video search for physical AI applications.
• Fluent in agentic AI workflows from prototyping to production integration.
• Clear communication skills and proven teamwork in dynamic, product-focused environments.