Prompt Engineering from scratch

What you'll learn

By the end of this course you will walk away with three real things:

1. A clear intuition of how an LLM “thinks” — so you stop prompting blind and start designing on purpose.
2. A toolbox of 6 prompting patterns that work on any model: zero-shot, few-shot, role prompting, chain-of-thought, JSON prefill, and self-reflection.
3. A deployable mini-project: a data extractor that takes free-form text and returns reliable structured JSON, with tests.

Every lesson is fully interactive: code runs in your browser and you can fix mistakes on the fly. No passive videos, no context-free theory.

Before we start

You do not need to know AI. You do need:

• To read JavaScript or Python.
• A free account (you create it in lesson 1).
• A willingness to be wrong. The best prompts come from iterating.