Mcp Crash Course: Create A Conversational Multi-Agent System

Build intelligent LLM agents that interact with tools, databases, and users using the MCP protocol

What you'll learn

Learn how to build intelligent agents that use LLMs to make decisions and perform actions in real time

Use the MCP protocol to integrate language models with tools, data, and external systems

Create custom Tools that LLMs can use to interact with their environment

Develop MCP servers and clients capable of asynchronous and efficient communication

Build conversational interfaces with Streamlit connected to LLM-based agents

Learn through practical examples how to build, test, and expand your own connected LLM agents

Requirements

Basic knowledge of Python. OpenAI API key required.

Description

Welcome to our course: MCP Crash Course: Create a Conversational Multi-Agent System.Model Context Protocol (MCP) is a modern, lightweight protocol designed to orchestrate interactions between LLMs and tools with security, control, and clarity. Have you ever imagined building LLM-based agents (Large Language Models) that truly interact with the world around them? With the course "MCP Crash Course: Create a Conversational Multi-Agent System", you’ll learn how to build agents that don’t just generate text — they take action, query databases, access external tools, and interact with users in a structured and useful way.This course is designed for AI professionals and enthusiasts who want to go beyond the basics and explore practical integrations between language models and real-world systems. The foundation for all this is MCP.You’ll start with the fundamentals of agents — understanding how they work, how they “think,” and how they make decisions. Then, you’ll dive into the core concepts of MCP: Tools, Resources, and Prompts, and see how they enable the creation of agents with real capabilities for reading, writing, and reasoning.The hands-on portion is the heart of the course: you’ll build real MCP clients and servers, create interfaces with Streamlit, and connect everything to an LLM to develop a powerful, fully functional conversational chatbot — one that retrieves data, responds intelligently, and executes commands with precision.If you're looking for autonomy, technical clarity, and real-world AI applications, this course is for you.Welcome aboard!

Overview

Section 1: Introduction

Lecture 1 Introduction

Lecture 2 MCP Fundamentals

Lecture 3 Prerequisites

Lecture 4 Course Files Download

Section 2: Create a Basic MCP Application: Client and Server

Lecture 5 Environment Setup

Lecture 6 Creating the MCP Server Application

Lecture 7 Creating the MCP Client Application

Lecture 8 Testing with Different Protocols

Section 3: Create a Conversational Multi-Agent System with MCP

Lecture 9 Multi-Agent MCP Application Fundamentals

Lecture 10 Database Schema for Our Application

Lecture 11 Creating the Database

Lecture 12 Running SQL Queries

Lecture 13 Environment Setup

Lecture 14 Creating the MCP Server Application

Lecture 15 Building Tools for the Server

Lecture 16 The Basics of the Client Application

Lecture 17 Session Management

Lecture 18 Creating MCP Agents

Lecture 19 Implementing the Chat Resolution Function

Lecture 20 Testing the Application: Case 1: Customer Interested in Vehicles

Lecture 21 Testing the Application: Case 2: Maintenance Schedule

Lecture 22 Testing the Application: Case 3: Scheduling a Test Drive

For anyone interested in learning how to build agents with MCP in practice.