welcome, human.

Files, folders, apps, paths — the things every later lesson assumes you already know.

Navigate, create, and manage files like a developer.

Install Claude Code, say hi in your terminal, and learn what it actually is — and isn't — before you trust it with anything bigger.

A short vocabulary tour — client and server, API, database, frontend and backend — so the rest of the course doesn't drown you in jargon.

Point Claude at a real codebase and start asking real questions. Files, functions, errors, history — by the end you can open anything and feel oriented, without writing a line.

Your first builds with Claude — point it at your own files, make real changes to real code, read what it proposes, push back when it's almost right, and back out cleanly when something goes wrong.

Read the prompt, judge the blast, pick the mode, and know what brings you back when something goes wrong.

When one prompt isn't enough — plan before you act, keep long work legible with TODOs, and steer Claude when it starts drifting instead of letting it run aground.

Land your work for real — commit it, branch it, push it, open a pull request, and handle review feedback, with Claude running the git and you holding the judgment.

Personalize Claude Code so it knows your project, remembers your preferences, exposes your common workflows as commands, and stops asking permission for the safe stuff.

Stop being the bottleneck. Hand research to subagents, package your repeating workflows as skills, and run several Claudes at once — all in a single sitting.

The last level. Plug Claude into your other tools, set up automations that run without being asked, and let Claude work on a schedule or on every pull request — long after you've closed the laptop.