What is Pi?

Pi is a free, open-source agent harness that you install on your personal computer. You interact with it from your terminal, having a back-and-forth dialogue with an AI model, much like you would with ChatGPT or Claude. But instead of just chatting, Pi can read and write files on your computer, write code, run commands, and interact with other applications.

Despite being built around a terminal, Pi is not just a tool for writing software. You can use it to automate any task on your computer, from managing files to controlling other applications. It's like having an assistant that can understand your intentions, help you brainstorm, do deep research, and build creative things.

Pi works with 15+ AI model providers, including Anthropic (Claude), OpenAI (GPT), and Google (Gemini). Rather than baking in features like sub-agents, plan mode, or MCP support, Pi keeps its core minimal and lets you extend it yourself with extensions, skills, and prompt templates — or install a package that does it your way.

Who is this for?

This course is for anyone who wants to expand their creative potential with AI. Whether you're a student, a hobbyist, a professional developer, or just curious about AI, Pi can help you do more with less effort.

Pi is terminal-based, so this course expects some comfort with a command line. You don't need to be a programmer — just willing to type commands and follow along.

This course is also for experienced developers. It'll get you set up with a good foundation for using Pi effectively, and you'll learn some tips and tricks for getting the most out of it.

What you'll learn

By the end of this course, you'll know how to:

• Install and configure Pi
• Use AI models to help you write and edit code
• Understand Pi's trust model and how it differs from other agents
• Extend Pi with skills, prompt templates, and extensions
• Navigate and branch your session history
• Build websites, games, and other creative projects!

How it works

The first lesson — Installation — is done here in the browser. Once you have Pi up and running, the rest of the course happens interactively inside Pi itself. Each lesson page has a prompt you copy and paste into Pi. The agent reads the lesson, works through the material with you, and marks it complete — which syncs back to this page in real time.