Cisco has introduced a new solution for building AI infrastructure, aimed at powering workload data fabrics and enabling enterprises to deploy agentic AI securely at scale. The Cisco Secure AI Factory with NVIDIA now expands to new use cases, including accelerating retrieval-augmented generation (RAG) pipelines for faster and safer data access.
At the core of the solution are Cisco AI PODs, now integrated with VAST InsightEngine, a key feature of VAST Data’s AI OS. These AI PODs deliver an end-to-end architecture using the NVIDIA AI Data Platform reference design, transforming raw enterprise data into AI-ready datasets. Built with Cisco UCS servers and NVIDIA RTX PRO 6000 Blackwell GPUs, the system offers high-performance computing for next-generation AI applications, while Cisco’s high-speed ethernet networking ensures seamless connectivity between compute and data.
This unified architecture supports near-real-time AI responses, enabling enterprises to reduce RAG pipeline latency from minutes to seconds. It allows AI agents to operate continuously, learn dynamically, and provide contextual insights at scale, with built-in governance and role-based access controls to safeguard sensitive data.
Cisco executives highlighted the shift from basic chatbots to enterprise-ready AI agents capable of solving real business challenges. “We are designing the architecture for how the enterprise will build the next generation of AI factories,” said Jeremy Foster, senior vice president and general manager, Cisco Compute. NVIDIA’s Justin Boitano added that the next wave of AI will rely on enterprise data for precise, up-to-date insights, while VAST Data emphasized its milestone role in delivering the first integrated design for RAG acceleration at scale.
The Cisco AI POD with VAST InsightEngine and NVIDIA AI Data Platform is now orderable, representing the first in a series of AI services PODs designed to meet the growing demand for enterprise AI use cases.
