Interactive Prompt Engineering Tutorial

What it does

• Nine chapters in beginner→advanced order: prompt structure, clarity, roles, separating data from instructions, output formatting, thinking step by step, examples, hallucinations, complex prompts
• An appendix beyond standard prompting: chaining prompts, tool use, search and retrieval
• Every lesson ends with an Example Playground to rerun the lesson's prompts, and a separate answer key is linked from the README
• Two ways to work through it: the notebooks in this repo (Anthropic 1P folder) or the same course as a Google Sheets workbook — the README recommends the Sheets version; an AmazonBedrock folder mirrors the course for Bedrock credentials

When to use it

Hands-on learners who want guided practice with exercises rather than reference prose; also the fastest way to internalize Anthropic's own mental model of what makes prompts work, chapter by chapter.

Getting started

# prerequisite: an Anthropic API key; the notebooks pip-install the anthropic SDK inline
git clone https://github.com/anthropics/prompt-eng-interactive-tutorial.git
cd "prompt-eng-interactive-tutorial/Anthropic 1P"
jupyter notebook 00_Tutorial_How-To.ipynb

Things to know

The README still says the tutorial uses Claude 3 Haiku and names Sonnet and Opus as the other options — the exercises predate the current model lineup, so treat model-specific behavior as dated even though the techniques transfer. Work in chapter order as the README instructs; the exercises build on earlier chapters.