Multi Agent AI and Network Knowledge Graphs for Change
summary
A production pipeline links ServiceNow tickets to a five‑agent system that consults a layered network knowledge graph implemented on ArangoDB using OpenConfig schema as a digital twin, pulls a config PR from GitHub, runs deterministic tests (Batfish/routenet style) and writes test reports back into the ITSM ticket. A fine‑tuned query agent replaces token‑heavy AQL loops, cutting token use and latency drastically, and the executor agent snapshots live graph state, runs tests, flags failures and emits remediation suggestions tied to ticket approvals.
AI-written summary · about
chapters (8)
- 0:00Introduction and problem
- 2:31System architecture overview
- 3:39Knowledge graph requirements
- 5:17Graph database selection
- 6:35Knowledge graph design
- 7:22Agentic framework and agents
- 13:05System demonstration
- 17:44Evaluation and conclusion
related talks
14mKnowledge Graphs as the Backbone of AI-First Software Delivery
43mArchitectural Blueprints for AI Agents
28mAgentic GraphRAG: Multi-Agent Knowledge Graph
29mAgentic GraphRAG: Multi-Agent Knowledge Graph Construction for Research Teams
16mMulti-Agent Intelligence for Next-Gen Insights
29mTransparent Multi-Agent AI
15mGraphRAG-Powered AI Agents: Transforming Corporate Metadata into Intelligent Knowledge
17mAI TechOps: Data and Infra for AI Foundation to Scale and Manage 1000s of Cohesive Agents