VAST Data unveils Foundation Stacks for NVIDIA AI OS

VAST Data has launched Foundation Stacks, an open source library that turns selected NVIDIA AI Blueprints into deployable pipeline implementations that run on the VAST AI Operating System.

NVIDIA AI Blueprints are reference workflows for building AI applications and agents using NVIDIA AI Enterprise software. Often used as starting points for prototypes and early deployments, they typically require additional work before they can run in production.

That work often includes integrating infrastructure, orchestration, and data services, along with establishing security and operational controls. VAST positions Foundation Stacks as a way to package these building blocks into repeatable templates that run natively on its platform.

Production packaging

Foundation Stacks extend NVIDIA AI Blueprints into what VAST describes as production-ready implementations, letting developers focus on application logic rather than assembling platform components. The library targets deployments where data access, database services, compute orchestration, eventing, and pipeline execution run in a single environment.

The templates can be deployed wherever the VAST AI OS runs, including cloud environments and on-premises installations. VAST also pointed to its newly announced CNode-X platforms and the NVIDIA AI Data Platform reference design as part of the broader deployment picture.

The announcement adds to a growing set of tools vendors are offering for “agentic” systems and pipeline-based AI workloads. Many enterprises are running pilots across functions such as customer support, research, and document processing, but operational work to connect models to data and run workflows reliably remains a common constraint.

First templates

The initial release includes two Foundation Stacks based on NVIDIA AI Blueprints for Video Search and Summarisation (VSS) and NVIDIA AI-Q.

The VSS-based stack targets video intelligence workloads. It can ingest large volumes of live or archived video and supports semantic indexing, summarisation, and interactive question-and-answer experiences. The stack uses data and pipeline services that run on the VAST AI Operating System.

The AI-Q-based stack targets research-style agent workloads over private enterprise data sources. It provides a base for building custom “AI researchers” and includes persistent context, secure context handling, reasoning pipelines, and trusted agent execution on the VAST AI OS.

VAST framed the stacks as a bridge between blueprint examples and operational deployments. “NVIDIA AI Blueprints have given the market an important starting point for building next-generation applications, but enterprises still need a production-ready way to deploy and operate those capabilities at scale,” said John Mao, Vice President, Global Technology Alliances at VAST Data.

Mao added: “With VAST Foundation Stacks, VAST is taking the architectural patterns behind leading NVIDIA Blueprints and giving customers a faster path from experimentation to production for scalable AI pipelines, video intelligence, and agentic AI systems.”

Data readiness

NVIDIA used the announcement to highlight the complexity of preparing enterprise data for AI systems that operate continuously. “As enterprises transition to production AI at scale, preparing enterprise data for AI has become one of the biggest challenges,” said Adel El Hallak, Vice President, Product at NVIDIA.

“Turning data into AI-ready pipelines needs to be done continuously and requires full-stack acceleration across compute, networking and software. By extending NVIDIA AI Blueprints with the VAST AI Operating System, customers can prepare and serve data so intelligent agents are always working off the most recent and accurate information,” said El Hallak.

The two companies have been working on reference patterns that more closely link compute and data services. That work is reflected in NVIDIA's AI Data Platform positioning, which describes an approach to data infrastructure that supports AI workloads across on-premises and cloud footprints.

Foundation Stacks will be distributed through a public GitHub repository. VAST also plans interactive demos and sandbox environments for customers and partners.

Additional Foundation Stacks are scheduled for release in the coming months, including templates focused on specific industries.