Getting Started with Claude Code

Claude Code is the most capable way to put AI to work on real tasks — not chatting about work, but doing it: reading files, running commands, writing code, and finishing jobs end to end. This course takes you from zero to your first shipped automation.

What you’ll be able to do

• Install and authenticate Claude Code on Mac or Windows
• Give it safe access to a real project
• Run your first multi-step task and understand what it’s doing
• Build one automation you’ll actually keep using

Module 1 — Setup without the footguns

Install the CLI, sign in, and learn the permission model before you need it. Claude Code asks before it acts; understanding what it asks — file edits, shell commands, network calls — is the difference between confidence and anxiety. We set up a sandbox folder first, so nothing it does can touch anything you care about until you’re ready.

Module 2 — Your first real task

Forget toy prompts. We hand Claude Code a messy, real job — “organize this folder of invoices and rename everything by date and vendor” — and watch how it plans, asks, and executes. You’ll learn the two habits that matter most: scoping the task tightly, and reviewing the plan before saying yes.

Module 3 — From task to automation

A task you run twice is an automation waiting to happen. We turn Module 2’s cleanup into a repeatable script Claude Code maintains for you, and cover when to keep a human in the loop (spoiler: anything that sends, deletes, or spends).

Module 4 — Where this goes next

Claude Code can manage projects, not just tasks: connected tools, scheduled runs, memory between sessions. We close with a map of what’s possible so you know which direction to grow.

The practitioner rule that carries this whole course: never automate something you haven’t done manually at least once. You can’t review what you don’t understand.