AI for Network Operations: From Traffic Spike to Root Cause to Fix (Kentik AI Advisor Demo)

Less than 10% of companies are AI-ready, according to Gartner — yet network teams are being told to "do something with AI in 2026." So what does that actually mean for network operations? In this 28-minute session, Kentik's Philip Gervasi (Director of Technical Evangelism) and Sean McGinley (Solutions Architect) walk through the progression from traditional network visibility to observability to network intelligence, explain how Kentik's AI Advisor agent system works under the hood, and demonstrate a live troubleshooting workflow on an Azure virtual hub that goes from traffic spike detection to root cause to cost optimization to CLI configuration — all through natural language. This isn't a chatbot bolted onto a dashboard. AI Advisor is an agent system that accesses real-time and historical telemetry (NetFlow, SNMP, streaming telemetry, syslogs, cloud logs), uses retrieval augmented generation (RAG) to ground answers in your actual network data, and builds step-by-step reasoning plans to perform root cause analysis. The session covers the architecture (LLM selection, tool calling, RAG pipelines, security boundaries), the concept of AI reasoning for network operations, and practical capabilities including DDoS vulnerability assessment, runbook-based troubleshooting, and automated cost optimization recommendations. What you'll learn: Why less than 10% of companies are AI-ready and what that means for network teams The progression from network visibility to observability to network intelligence How Kentik AI Advisor's agent architecture works (LLMs, RAG, tool calling, security) How AI reasoning creates step-by-step troubleshooting plans from natural language questions How to capture institutional knowledge in natural language runbooks Live demo: detecting an Azure Virtual Hub traffic spike, diagnosing inefficient routing, and getting VNet peering CLI commands Learn more about Kentik AI Advisor: https://www.kentik.com/product/ai-adv... ─── RELATED RESOURCES ─── 📊 More AI and network operations content from KentikHQ: How to Troubleshoot BGP Faster with Kentik AI Advisor:    • How to Troubleshoot BGP Faster with Kentik...   Troubleshooting SD-WAN with Kentik Journeys AI:    • Troubleshooting SD-WAN with Kentik Journey...   AI-Assisted Network Monitoring: Real or Hype?:    • Assisted Network Monitoring: Real or Hype?...   AI NetOps: How AI and Machine Learning Transform Network Operations:    • AI NetOps: How AI and Machine Learning Tra...   📊 Related demos: Kentik Data Explorer Tutorial (Network Traffic Analysis):    • Kentik Data Explorer Tutorial: Network Tra...   How to Analyze Network Traffic Costs:    • How to Find and Reduce Network Traffic Cos...   🔗 Learn more on kentik.com: Kentik AI Advisor: https://www.kentik.com/product/ai-adv... Kentipedia: https://www.kentik.com/kentipedia/ ─── ABOUT KENTIK ─── Kentik is the network intelligence platform for modern infrastructure teams. Kentik provides network observability, DDoS protection, cloud cost optimization, BGP monitoring, and AI-powered network analytics for enterprise, cloud, and service provider networks. 🔗 Start a free trial: https://www.kentik.com/go/get-started/ 🔗 Subscribe:    / @kentikhq   #AINetOps #NetworkIntelligence #AIAdvisor #Kentik #NetworkOperations #Azure #CloudCost #NetworkAutomation CHAPTERS: 0:00 Introduction: AI for Network Operations with Philip Gervasi and Sean McGinley 1:30 Why Network Teams Struggle: Manual Clue Chaining and Institutional Knowledge Gaps 4:09 The Roadblocks to Implementing AI in Network Operations 5:36 Gartner: Less Than 10% of Companies Are AI-Ready 8:20 From Visibility to Observability to Network Intelligence 9:11 Not All AI Is Created Equal: What AI for NetOps Really Means 11:22 Introducing Kentik AI Advisor: An Agent System Built for Network Operations 12:03 How AI Advisor Works: Agent Architecture, LLMs, RAG, and Tool Calling 16:59 AI Reasoning Explained: How the System Builds Step-by-Step Troubleshooting Plans 19:30 AI Advisor Capabilities: Cost Optimization, Troubleshooting, DDoS Assessment 20:55 Natural Language Runbooks: Capturing Institutional Knowledge 23:16 Live Demo: Azure Virtual Hub Traffic Spike Detection and Investigation 24:13 Asking AI Advisor: Show Me Traffic Over My Azure Virtual Hub 24:49 Cost Optimization: How Inefficient Routing Drives Up Azure Costs 25:38 Getting CLI Configuration: How to Set Up VNet Peering via Azure CLI 27:49 Conclusion: AI-Augmented Network Operations in Practice