RAG for Networks: Building AI-Ready Knowledge Bases
Retrieval-Augmented Generation (RAG) is becoming a cornerstone of enterprise AI adoption, but how can it be applied effectively in the world of networking? In this session, we’ll define RAG in plain terms and show how it can deliver tangible business benefits by transforming fragmented network documentation, design records and network procedures into an AI-ready knowledge base. We’ll explore why RAG is particularly relevant for networking and how it differs from MCP-style orchestration and Agentic actions. Attendees will learn why traditional text chunking and vector databases, while useful, are not enough on their own. Instead, we’ll demonstrate how a graph-aware format like JSON-LD can supercharge retrieval precision, enabling GenAI systems to reason over complex network relationships. By the end of the session, participants will understand the building blocks of effective network knowledge base systems—JSON-LD, vector databases, and chunking strategies—and how they combine to make RAG solutions more powerful, reliable, and actionable for real-world networking challenges.
Speaker
Gary Woodward N/A
Network Automation and AI Solution Specialist
Delivering courses and Tech Sessions on the PacketCoders platform, covering a wide range of popular Network Automation topics. With extensive experience across both the SaaS and Telco industries, I have worked at the intersection of software and networking for many years. Today, my focus is on bridging Network Automation and AI to deliver meaningful business outcomes.... read more