Cursarium logoCursarium
beginnerFree

Anthropic's Prompt Engineering Interactive Tutorial

by Anthropic · Anthropic

9 chaptersUpdated 2026-06

Our Verdict

Worth taking

Anthropic's Prompt Engineering Interactive Tutorial is a free, open-source course (CC BY-NC 4.0) that teaches prompt engineering for Claude through nine structured chapters and an advanced appendix. It is genuinely hands-on: each lesson includes an example playground where you edit prompts and immediately see how Claude's responses change, delivered as Jupyter notebooks or, more accessibly, a Google Sheets version. With 36k+ GitHub stars it is one of the most widely used free prompt engineering resources, and being authored by the model provider makes the techniques authoritative. The trade-off is that it teaches against Claude's API (it uses the low-cost Claude 3 Haiku in examples) and is a self-guided repository rather than a polished video course with a certificate. An excellent, low-cost way to actually practice prompting, best for those willing to run code or a spreadsheet.

Free, open-source, and unusually practical: it lets you actually write and troubleshoot prompts against Claude with first-party guidance, which is more valuable than passive reading for learning prompt engineering.

Best for: Developers, technical practitioners, and motivated learners who want hands-on, structured practice writing and debugging prompts for Claude, and who are comfortable running a Jupyter notebook or working in a Google Sheet.

Skip if: Learners who want a polished video course, a completion certificate, or a fully provider-neutral overview; also less ideal for those unwilling to set up notebooks or an API key.

About This Course

Anthropic's free, open-source interactive tutorial that teaches prompt engineering for Claude step by step across nine chapters plus an advanced appendix, with hands-on exercises in Jupyter notebooks (or a Google Sheets version).

What You'll Learn

How to structure basic prompts and write clear, direct instructions for Claude
How to assign roles and separate data from instructions to improve reliability
How to format outputs, speak for Claude, and use step-by-step ('precognition') reasoning
How to use examples (few-shot prompting) effectively
Techniques to reduce hallucinations and increase answer accuracy
How to build complex, production-style prompts for real industry use cases (chatbots, legal, financial, coding)
Advanced appendix topics: prompt chaining, tool use, and search & retrieval

Curriculum

Beginner: Basic Prompt Structure, Being Clear and Direct, Assigning Roles

Chapters 1-3 cover prompt fundamentals.

Intermediate: Separating Data from Instructions, Formatting Output & Speaking for Claude, Precognition, Using Examples

Chapters 4-7 cover core reliability and quality techniques.

Advanced: Avoiding Hallucinations, Building Complex Prompts

Chapters 8-9, including industry use cases for chatbots, legal, financial, and coding.

Appendix: Chaining Prompts, Tool Use, Search & Retrieval

Optional advanced techniques beyond the core chapters.

Prerequisites

  • Basic comfort running Jupyter notebooks (or using the Google Sheets version)
  • An Anthropic API key to run the live examples (the tutorial uses the low-cost Claude 3 Haiku)
  • No prior prompt engineering experience required

Instructor

Anthropic

Instructor · Anthropic

Pros & Cons

Pros

  • Free and open-source (CC BY-NC 4.0), authored by Anthropic itself
  • Truly hands-on: an example playground in every lesson for live prompt experimentation
  • Very widely used and vetted (36k+ GitHub stars)
  • Spans beginner to advanced, including real industry use cases and an advanced appendix

Cons

  • Self-guided repository, not a polished course; no video lectures and no certificate
  • Requires running Jupyter notebooks (or a Google Sheet) and an Anthropic API key to do the exercises
  • Examples are Claude-specific and use older Claude 3 Haiku, so some model details lag current models
  • CC BY-NC 4.0 license restricts commercial reuse of the materials

Alternatives To Consider

How we reviewed this course

This is an independent editorial assessment by Cursarium, based on Anthropic's published course materials and aggregated public learner feedback (last reviewed 2026-06). We have not independently completed the course. Links to providers are standard references, not paid placements.