🎙️ Welcome to the_context(), a weekly show about Generative AI where we discuss latest news and show technical demos! In this episode, we deep dive into the Model Context Protocol (MCP) with lambda functions. Mike walks us through how MCP works and how you can create MCP servers with just a few lines of code—opening the door to simpler, more modular, and more scalable AI agent integrations. ---- 👥 Speakers: - Mike Chambers | Developer Advocate @AWS - Tiffany Souterre | Developer Advocate @AWS 💻 Resources: - MCP Guide: - MCP SDK: - Lambda MCP Server Demo: - AWS Lambda MCP Server on awslabs: And the test client that Mike built uses Amazon Nova Pro, find out more about this mode here: 👇 Got questions or want to see a specific demo in a future episode? Let us know in the comments! #GenerativeAI #MCP #AWSLambda #AIAgents #TheContextPodcast #DeveloperTools #bedrock 0:00 – Intro to MCP and What It Solves 1:18 – Decoupling Agents and Tools with MCP 2:50 – Natural Language Tool Descriptions and Protocol Benefits 4:06 – Transport Evolution: From STDIO to Streamable HTTP 6:44 – Simpler Server Implementation Using HTTP 8:05 – TypeScript SDK Adds Streamable HTTP Support 9:47 – Creating a Lambda-Based MCP Server 11:56 – Session Management with DynamoDB and SAM 15:00 – Tool Examples and Natural Language Metadata 17:28 – Bearer Token Authorization for API Gateway 19:48 – CloudFormation Deployment and Inspector Demo 23:15 – CLI Docker Client for MCP Server Testing 27:03 – Comparison with AWS Labs Lambda MCP Project 30:03 – Wrap-Up and Invitation to Explore Further