Building enterprise AI agents with Model Context Protocol

The world is moving beyond simple AI chatbots to agentic AI, or autonomous systems that use reasoning to perform complex tasks by interacting with external tools. But connecting large language models (LLMs) to tools across an enterprise traditionally requires complex, custom code for every integration. This is where the Model Context Protocol, or MCP, can unlock the business potential of agentic AI. MCP is an open-source standard that connects LLMs to outside data and tools.

Principal Product Manager Peter Double and Senior Technical Marketing Manager Cedric Clyburn unpack how MCP brings portability, efficiency, and governance to AI agents, and how Red Hat is building a path for engineers to innovate faster with a full lifecycle approach.

00:00 Introduction
01:03 The Evolution of Interacting with LLMs
03:38 The Shift to Tool-Calling and Agentic AI
05:42 Introducing the Model Context Protocol (MCP)
07:23 Closing the Gaps: Making MCP Enterprise-Grade
08:14 Red Hat’s Response: Security, Governance & Scale
09:25 Model Context Protocol (MCP) in Red Hat OpenShift AI v3.0
10:50 Red Hat’s Full Approach to MCP & End-to-End Experience
12:14 Demo 1: Using MCP Through a Chat Client (Kubernetes & Slack)
17:53 Demo 2: MCP Integrated in a Blackjack AI Application
19:42 Future roadmap and closing

🧠 Explore Agentic AI with Red Hat → https://www.redhat.com/en/products/ai/agentic-ai

#RedHat #AgenticAI #AIEngineering #MCP