Handle
Connecting…
Home

Workshop Modules

Build an AI agent from scratch, step by step. Each module builds on the previous one — by the end you'll have a complete AI agent connected to real-world tools.

10 modules · ~3 hours total Python · C# · TypeScript · Go · Java

Getting Started

01

What are AI Agents & MCP?

~15 min

Understand the big picture: what AI agents are, why they need tools, and how MCP connects them.

Start module
02

Environment Setup

~15 min

Install Docker, get API keys, clone the repo, and verify everything works.

Start module
03

Explore the Running System

~15 min

Poke around the existing system with curl. See how tools/list and tools/call work.

Start module

Build It

04

Build the MCP Server

~30 min

Create an MCP-compliant server from scratch with JSON-RPC 2.0, tool discovery, and a weather tool.

Start module
05

Build the AI Agent

~30 min

Build an AI agent that discovers tools dynamically, talks to an LLM, and orchestrates tool calls.

Start module
06

Wire Agent ↔ MCP Server

~15 min

Connect your agent and MCP server with Docker Compose and run the full system end-to-end.

Start module

Extend It

07

Add a Custom Tool

~20 min

Add a random facts tool to the MCP server. The agent discovers it automatically.

Start module
08

Connect to Real APIs

~20 min

Add a news/RSS tool that pulls real data from external APIs.

Start module

Advanced

09

Run with Local Models (Ollama)

~20 min

Swap OpenAI for a local LLM using Ollama. No API key needed, runs on your machine.

Start module
10

Make It Bulletproof

~20 min

Add error handling, debugging, context management, and production hardening.

Start module

Bonus Challenges