Building AI Agents — Designing and Building AI That Acts on Its Own
A practical, developer-focused course on designing and building AI agents that, given a goal, work out the steps themselves, use tools, and carry out multiple steps. All 7 chapters: how they work, your first agent, MCP integration, multi-agent, evaluation, safety, and production operations.
About this course
Who this course is for
- Developers who want to move past chatbots and let AI act on its own
- You want to understand tool integration and MCP as mechanisms, not just by feel
- You want evaluation and safeguards settled before an agent goes to production
What you will be able to do
- Design the structure an agent uses to cycle through goal, plan, execution and verification
- Connect external tools through MCP and split the work across multiple agents
- Head off runaway behavior, infinite loops and cost overruns at design time
Prerequisites
Easier to follow if you have called an API before. LLM fundamentals are covered in the text.
Suggested pace
Seven chapters, 15 to 25 minutes each. Expect two to three weeks if you build as you read.
Table of Contents
1
Chapter 1
What Is an AI Agent — How It Works and How It Differs from RPA and Chatbots
2
Chapter 2
Build Your First Agent — A Minimal Setup and the Loop
3
Chapter 3
MCP and Tool Integration — Giving Your Agent Tools
4
Chapter 4
Multi-Agent Design and A2A
5
Chapter 5
Evaluation (Evals) and Observability
6
Chapter 6
Guardrails and Security
7
Chapter 7
Frameworks and Production
Log in to track your progress. Log In