Course

AI Engineering

From what the model actually sees to how teams ship agents safely. Twenty-five short lessons, each with a picture, an interactive lab and a check. Choose the right architecture instead of defaulting to “just prompt it.”

0 of 25 lessons done

This browser is not letting the page save anything (private window or blocked site data?), so progress will not be remembered.

How it works

• Go in order. Each lesson builds on the one before. Every page ends with links to the previous and next lesson.
• Pass the check. Each lesson ends with a short quiz. Pass it and the lesson is marked done. You can also mark a lesson done yourself.
• Progress stays in this browser. Nothing is sent anywhere. To move to another device, download your progress below and load it there.
• Three lessons have code. 0021, 0023 and 0024 are TypeScript exercises that run locally against a scripted model, using Node 22 or later. No API key and nothing to install.

Cheat sheets

One-page summaries to come back to after a lesson.

• Map of the era
• Tokens
• Context window
• Pack, read, pick
• Find · paste · pick (RAG)
• Tools
• Two architectures
• Harnesses
• Skills
• Context engineering
• Just-in-time
• Sub-agents
• Loop engineering
• Graph engineering
• Evals
• Team floor
• Stack choice
• Plugins
• Production traces
• Eval suites
• Agent security
• Software factory

Your progress

Saved only in this browser. Download a copy to keep it, or to carry on from another device.