VAST Data has released Foundation Stacks, an open source library that extends Nvidia’s AI Blueprints into production-ready pipeline templates for enterprise deployments.
VAST Data, an AI infrastructure software company, announced Sunday the release of VAST Foundation Stacks, an open source library designed to bridge the gap between Nvidia’s AI Blueprint reference implementations and the production-ready deployments that enterprises actually need.
The announcement was made at Nvidia’s GTC 2026 conference in San Jose.
Nvidia AI Blueprints give developers a starting point for building AI applications and intelligent agents, providing templates that can be rapidly prototyped and customized for specific use cases. But most of those reference implementations still require significant integration work — stitching together infrastructure, orchestration layers and data services — before they can run securely and reliably at scale. VAST Foundation Stacks are designed to close that gap, extending the blueprints into repeatable, enterprise-ready templates that run natively on VAST’s AI Operating System.
By consolidating data access, database services, compute orchestration and pipeline execution into a single environment, the company said, organizations can deploy scalable AI pipelines without assembling complex infrastructure from scratch. The stacks can be deployed in cloud environments and on premises.
Also Read: “The AI Economy Has Moved From the Training Phase Into the Inference Phase,” Says Jensen Huang
The first two Foundation Stacks are built on Nvidia AI Blueprints for video search and summarization and for AI research workflows. The video-focused stack allows organizations to ingest large volumes of live or archived footage and extract insights through semantic indexing, summarization and question-and-answer capabilities. The research-focused stack provides a foundation for building AI systems that can work across private enterprise data sources, synthesizing hours of research in minutes.
“Enterprises still need a production-ready way to deploy and operate these capabilities at scale,” said John Mao, vice president of global technology alliances at VAST Data. “With Foundation Stacks, we are giving customers a faster path from experimentation to production for scalable AI pipelines, video intelligence and agentic AI systems.”
Adel El Hallak, vice president of product at Nvidia, said preparing enterprise data for AI had become one of the biggest obstacles to production deployment. “Turning data into AI-ready pipelines needs to be done continuously and requires full-stack acceleration across compute, networking and software,” he said. “By extending Nvidia AI Blueprints with the VAST AI Operating System, customers can ensure intelligent agents are always working from the most recent and accurate information.”
Also Read: Regulated Industries Are Rewriting Their AI Architecture
VAST said it plans to release additional Foundation Stacks in the coming months, including implementations focused on specific industries. The library will be available through a public GitHub repository, with interactive demonstrations and sandbox environments planned for customers and partners.
