Skip to main content

Exercises Overview

What You'll Build​

The Fellowship Exercises are a hands-on journey from building your first AI agent to orchestrating complex distributed systems.

Each exercise builds directly on the previous one. By the end, you'll have a complete, production-like system where:

  • Agents make intelligent decisions
  • MCP Servers expose capabilities and operations
  • Services coordinate asynchronously
  • Events flow through the system
  • Workflows automate your entire pipeline

The Five Exercises​

1️⃣ React Foundation​

Learn the ReAct pattern by building your first agent that reasons, acts, observes, and repeats.

2️⃣ Jenkins MCP Server​

Build your first Model Context Protocol server that exposes Jenkins to agents.

3️⃣ Watcher Agent​

Create an intelligent, multi-server agent that monitors and controls pipeline execution.

4️⃣ Gitea MCP Server​

Build an advanced MCP server for version control operations and Git workflows.

5️⃣ Grand Finale​

Integrate everything: Orchestrate five services, implement event sourcing, and build your first production system.

Key Concepts You'll Learn​

🧠 Agents (Exercise 1, 3)​

AI agents that reason about problems, make decisions, and take actions in the real world.

The ReAct pattern: Reason → Act → Observe → Repeat

🔧 Tools (Exercises 1, 2, 3, 4, 5)​

Functions that agents can call to interact with systems.

You'll learn to define, implement, and expose tools through MCP.

📡 Model Context Protocol (MCP) (Exercises 2, 4)​

A standard for agents to discover and use tools exposed by servers.

You'll build servers using the MCP standard.

🌊 Event Sourcing (Exercise 5)​

A pattern where changes are recorded as immutable events, replayed to build state.

You'll implement event-driven workflows with SQS.

🤝 Multi-System Orchestration (Exercise 5)​

Coordinating multiple services, tools, and tools to accomplish complex goals.

You'll tie everything together into one cohesive system.

How to Approach These Exercises​

✅ Do This​

  • Follow the exercises in order (1 → 2 → 3 → 4 → 5)
  • Read all the provided documentation and examples
  • Experiment! Try variations and see what happens
  • Ask questions on the exercise forums or with your instructor
  • Take breaks between exercises to absorb concepts

❌ Don't Do This​

  • Skip exercises (each depends on previous knowledge)
  • Copy-paste code without understanding it
  • Rush through exercises (slower is faster)
  • Hesitate to ask for help

Your Learning Journey​

Exercise 1: Learn agent basics
    ↓
Exercise 2: Learn to expose capabilities
    ↓
Exercise 3: Combine agents and servers
    ↓
Exercise 4: Advanced server patterns
    ↓
Exercise 5: Bring it all together 🎉
    ↓
You're ready to build production AI systems!

Tips for Success​

1. Read the Setup Section First​

Each exercise has a "Getting Started" section with setup instructions. Follow it exactly.

2. Understand Before Coding​

Read the "Concepts" section before writing any code. This shows why you're doing something.

3. Copy the Template​

Each exercise provides a starting template. Understand what's already there before adding code.

4. Run Tests Frequently​

Use the provided tests to validate your work. Run them after each step.

5. Read Example Solutions​

After completing an exercise, review the annotated example solution to see alternative approaches.

6. Debug Systematically​

  • Check the error message carefully
  • Look at the test output
  • Review your recent changes
  • Consult the troubleshooting guide

Common Questions​

Q: What if I'm stuck on an exercise?​

A: Check these in order:

  1. Re-read the exercise README section
  2. Review the "Common Issues" section
  3. Look at example solutions
  4. Check the Troubleshooting Guide
  5. Ask your instructor

Q: Can I skip an exercise?​

A: Not recommended. Each exercise teaches concepts needed for the next one. But if you've completed similar exercises in the past, you can move faster.

Q: How long will this take?​

A: 12-17 hours total, spread over 3 weeks. That's 1-2 hours per day. You can go slower or faster depending on your pace.

Q: What if I want more challenge?​

A: Each exercise has an "Go Further" section with advanced variants. Try those after completing the main exercise.

Q: Can I use different languages?​

A: Some exercises show Python and JavaScript. Pick one and stick with it for consistency within each exercise.

Ready to Begin?​

You've got this! 🚀

→ Start with Exercise 1: React Foundation

Or jump to a specific exercise: