You've made it through the foundations. You know what a computer is, what a file is, what the internet is, what code is, what a terminal is, and what an AI model is. That's already more than most people who use these things every day.
Now we can finally talk about Claude Code itself — and more importantly, why it's a fundamentally different way of working.
The pitch in one paragraph
Claude Code is an AI that lives in your terminal.You run a command, you point it at a folder, and you have a conversation. You describe what you want — "build me a website that does X," "fix the bug where Y happens," "explain what this code does" — and Claude reads your files, writes new ones, runs the commands needed to test things, and shows you everything it's doing along the way.
What this replaces (and what it doesn't)
For decades, doing anything technical meant either learning to write code yourself or hiring someone who could. Claude Code collapses a lot of that. You can describe a problem in plain English and watch a very capable, very fast collaborator solve it with you.
What it doesn't replace is your judgment. Claude can write a thousand lines of code in two minutes; it cannot tell you whether they're the right thousand lines. It can build the thing you described; it cannot decide whether the thing you described is worth building. That part is still you.
Why a terminal, not a website?
It might seem odd that the most modern AI tool deliberately lives in the oldest-looking interface on your computer. There's a reason. The terminal is where code already happens. By living there, Claude Code can:
- Read and edit the actual files on your computer, not copies.
- Run real commands — start servers, run tests, install dependencies — and see the real output.
- Compose with every other developer tool you already use (git, editors, build tools).
- Work on whatever language or framework you want, because nothing about it is locked to one ecosystem.
A chat interface in a browser can't do any of that. Or rather, it can pretend to, but you have to keep copying code back and forth manually. Claude Code skips all the manual.
What you'll be able to do
By the end of this curriculum, you should be comfortable:
- Spinning up a new project from scratch by describing it in paragraph form.
- Reading and modifying an existing project you've never seen before.
- Asking Claude to investigate a bug, propose a fix, and apply it — and knowing how to sanity-check the result.
- Wiring up Claude to your own tools and APIs so it can do things specific to your work.
- Running multiple Claude agents in parallel on different parts of a problem.
Where to from here
Module 1 is done. The next module — Software Basics — teaches the shared vocabulary every developer takes for granted: git, editors, package managers, environment variables, all the plumbing that surrounds code. You don't need to be an expert in any of it; you do need to know what each thing is when Claude mentions it.
If you're itching to install Claude Code right now, that's in Module 3. The boring middle module makes it stick.
- Claude Code is an AI agent in your terminal that can read, write, and run code on your computer.
- It changes the skill from "type code" to "describe what you want and steer."
- It lives in a terminal because that's where real code work happens.
- Module 2 covers the developer vocabulary you need. Module 3 is where we actually install and use Claude Code.