Operations | Monitoring | ITSM | DevOps | Cloud

In this hands-on workshop, we covered how to build your own MCP server from scratch and connect it to AI tools like Cursor IDE or Claude Desktop. The first half is a live coding session you can follow along with to set up an MCP server for Kubernetes troubleshooting. In the second half, we take you behind the scenes at Komodor to show how we built our MCP Server MVP: a powerful bridge between AI assistants and Kubernetes infrastructure. This is just part of the 'magic' that helps the Klaudia agentic AI technology power Komodor's AI SRE Platform.

You’ve seen the vision. You’ve built the blueprint. Now it’s time to put AI to work. In this fast-paced, tactical session, we’ll showcase real-world MSP AI implementations — what’s working, what’s not, and how to adapt these examples to your own business.

AI Agents & Model Context Protocol John Capobianco, Head of Developer Relations at Selector, dives deep into AI Agents and the Model Context Protocol (MCP). In this session, John demonstrates Selector MCP in action — running as a client-server, connecting multimodal inputs, and even talking to Selector using microphone + TTS audio via Gemini CLI. He also showcases Sebastian Maniak’s Claude Desktop integration, where Selector MCP powers a chatGPT-like UI for network engineers. A practical look at how MCP is transforming AI into a true digital co-worker.

Building Your First GenAI RAG Application John Capobianco, Head of Developer Relations at Selector, walks through a 6-step process for building your first GenAI RAG application. From foundational building blocks to the path toward full AI agents, RAG remains a powerful tool with huge ROI. Even in a world of autonomous agents and MCPs, RAG is still one of the best ways for network engineers and IT leaders to query dozens of sources and unlock real value.

Strategic Imperatives for Infrastructure Leaders in an AI-Enabled World William Collins, Director of Technical Evangelism at Itential, explores the strategic imperatives facing infrastructure leaders in today’s AI-enabled world. He unpacks the Gartner Hype Cycle, the true monetary costs of network downtime, and shows how Itential + Selector can close the loop on AIOps with autonomous agents and MCPs.