PyData Boston 2025

Tutorial Material

GitHub Repo Link: https://github.com/chuxinliu/PyData_Boston_2025_MCP_Server_Tutorial

Target Audience

Anyone who is curious about MCP and wants to build it on its own.

Prerequisites

• An intermediate understanding of Python.
• No prior experience with LLMOps is required.

Tutorial Outline (90 Minutes)

Part 1: Concepts & Setup (20 mins)

• What is the Model Context Protocol? We'll explain the MCP Host, Client, and Server architecture and a breakdown of the core concepts.

Part 2: Hands-On: Lifecycle and Tool Discovery 50 mins)

• Environment Setup: A quick tour of the provided GitHub repository, explaining the project structure, dependencies, and a simple MCP client we'll use for testing.
• We'll define a simple Python function and then build the MCP logic to expose it as a "tool," complete with a name, description, and a JSON Schema for its inputs.
• Implementing tools/call. We will build the endpoint that receives a tool name and arguments from the AI client, securely executes the corresponding Python function, and returns the result in the MCP-specified format.
• We'll run our completed server and use a provided command-line MCP client to simulate an AI Host. Attendees will see their client discover the custom tool via tools/list and then execute it via tools/call, seeing the live JSON-RPC exchange.

Part 3: Demo, Recap & The Road Ahead (20 mins)