Twenty minutes now. Everything works better after.
Most people skip this part. Don't. The difference between Claude feeling mediocre and Claude feeling like the most useful tool you've ever used comes down almost entirely to setup. Do this before you touch any chapter.
Pro tip: do this on your laptop or desktop
The setup process involves installing the desktop app, creating folders, connecting apps, and building skills. That's all much easier on a full-size screen. Use your phone for Claude later — but do the bootcamp on your computer.
What you'll do in setup
Get Claude set up
Create your account, download the desktop app, and pick the right plan.
Turn on Memory
So Claude remembers who you are across every conversation instead of starting from zero.
Bring your ChatGPT context
If you've been using ChatGPT, export your data and bring everything over so you don't start from scratch.
Set global instructions & response style
Standing rules that apply to every conversation — so Claude works the way you want from the start.
Quick orientation
A 30-second map of Chat, Projects, and the other tools you'll hear about — so you know what's what before your chapter.
Step 1 — Get Claude set up
You need two things: a Claude account on the web, and the desktop app. Both are free to install — you'll upgrade to Pro in a moment.
Get set up
Create your account: Go to claude.ai and sign up (or sign in if you already have one). This is your home base — where you'll chat, build Projects, and do the majority of this bootcamp.
Download the desktop app: Get the Claude desktop app for Mac or Windows from claude.ai/download. Same account, syncs automatically. The desktop app gives you a dedicated Claude window outside your browser — great for keeping Claude open while you work. You'll want both the browser and the desktop app.
Your plan options
Free plan — works for basic chat, but has daily usage limits and no access to Projects or connectors.
Claude Pro ($20/month) — this is where everything in this bootcamp actually lives. You get higher usage limits, Claude's most powerful models, the ability to connect Claude to your email (Gmail or Outlook), calendar, cloud storage, Slack, Notion, Salesforce, and 50+ other tools — plus Claude Projects, your own dedicated AI workspace built specifically for your job. This is the biggest unlock in the entire bootcamp and it only exists on Pro.
Know your models
What is a "model"? When you use Claude, there's an AI brain behind the scenes processing your messages and generating responses. That brain is called a model. Anthropic (the company that makes Claude) offers several different models — think of them like different engines you can put in the same car. They all power Claude, but some are faster, some are smarter, and some use more of your daily allowance than others. You get to choose which one you're using at any time.
Sonnet — This is your default. Use it for 95% of everything: emails, research, writing, analysis, brainstorming, daily tasks. It's fast, capable, and uses the least of your daily usage allowance. When in doubt, use Sonnet.
Opus — The premium model. Smarter, deeper reasoning, better at complex multi-step tasks — think negotiation strategy, long document analysis, nuanced writing where every word matters. The trade-off: it uses significantly more of your daily usage limit. Save it for when the task genuinely requires deeper thinking.
Haiku — The lightweight model. It's fast but noticeably less capable. You can ignore it entirely for the kind of work you'll do in this bootcamp.
How to switch models: When you open a conversation on claude.ai, look at the top of the chat window — you'll see a model name (like "Claude Sonnet" or "Claude Opus") displayed near the top. Click on that name and a dropdown menu will appear showing all available models. Just click the one you want to switch to. That's it — your next message will use the new model. You can switch back and forth anytime, even mid-conversation. In the Claude Desktop app, you'll find the same model selector at the top of the chat.
What about usage limits? On the Pro plan, you get a daily usage allowance — think of it like a bucket of credits that refills every 5 hours. Every message you send uses some credits. Sonnet uses the fewest credits per message, Opus uses the most. So if you use Opus for everything, you'll run out much faster and have to wait for your bucket to refill. That's why the recommendation is: Sonnet for daily work, Opus only when you need the extra brainpower.
One more thing: AI models keep getting smarter. By the time you're reading this, the model names might have changed — but the principle stays the same. There will always be a default model that's great for everyday work (better on usage) and a premium model that's better for complex tasks (worse on usage). Use the default for most things, upgrade when it matters.
Step 2 — Turn on Memory
Here's something most people never find — and it changes everything about how useful Claude is.
Without memory on, every single conversation starts completely fresh. Claude has no idea who you are, what your job is, how you like to communicate, or anything you've ever told it. You're starting from zero every time.
With memory on, Claude holds onto the important things you tell it and brings them into every future conversation automatically. It gets smarter about you every time you use it.
Turn it on right now
Go to claude.ai
Click your profile icon → Settings
Find Memory and switch it on
Open a new conversation and tell Claude about yourself — your job, your company, your communication style, what you're working toward
Say: "Remember all of this about me"
Done — Claude saves it and carries it into every conversation going forward
You can go back to Settings → Memory → Manage any time to see what it's saved, update it, or add more. The more you give it, the better it performs.
Step 3 — Coming from ChatGPT? Bring everything with you
If you've been using ChatGPT, don't start from zero. The context you've built up, your preferences, how you've trained it — bring it all over.
Export from ChatGPT
Open ChatGPT → click your profile picture → Settings
Go to Data Controls → Export Data
Hit Export — you'll get an email with a download link within a few minutes
Download it — it has your full conversation history inside
Bring it into Claude
Look through the export for conversations where you gave real context about yourself — your job, communication style, preferences
Start a new Claude conversation, upload the file or paste the relevant parts in
Say: "Extract everything here that tells you about who I am, how I work, and my preferences — then remember it"
If you had custom instructions in ChatGPT, paste them in and say: "Save the relevant parts of this to your memory about me"
You put time into building that context. Don't leave it behind.
Step 4 — Set your global instructions
This is one of the most overlooked settings in all of Claude — and it takes two minutes to set up.
Global instructions are rules you set once that Claude reads before it touches anything. Every conversation, every task, every session — these rules are always loaded. Without them, Claude is guessing about how you like to work every single time. Most people either skip this completely or leave it empty.
What are global instructions?
Think of them as standing orders for an employee on their first day. You don't want to repeat "don't use corporate jargon" or "always ask before deleting something" in every conversation. You say it once, and it sticks. That's what global instructions do.
How to set them up
Go to claude.ai → click your profile icon → Settings
Look for "Profile" or "Personalization"
Find "What personal preferences should Claude consider in responses?"
Paste in your instructions (see below) and save
Done for you — starter global instructions (copy and paste these)
Rules for every conversation: - Never use emdashes. Ditch all corporate jargon — write like a real person. - Always ask clarifying questions before starting any task. If something is vague, ask — don't guess. - Always confirm before creating, deleting, or overwriting any document or file. - When you're uncertain about something, stop and ask instead of making assumptions. - Keep files and folders visible and organized — avoid hidden folders whenever possible so I can easily review and edit everything. - First drafts should be 90% ready. I want to review and tweak, not rewrite from scratch. - Be direct. Skip the filler. No "Great question!" or "Absolutely!" — just get to the answer.
You'll customize these more later
These are good universal defaults. When you get into your specific chapter, you'll add role-specific rules on top of these (like how you want emails written, or how detailed your reports should be). For now, these prevent the most common annoyances right out of the gate.
Step 5 — Set your response style
This one's quick but makes a noticeable difference. Claude has built-in response styles you can set as your default so you're not constantly asking it to "be more concise" or "write more formally."
How to set it
Go to claude.ai → click your profile icon → Settings
Look for "Response Style" or "Preferred style"
Choose from the built-in options (Concise, Formal, Explanatory) or write your own custom style
If you're not sure, start with "Concise" — you can always ask Claude for more detail in the moment, but it's harder to un-do a wall of text
Step 6 — Quick orientation: what's what
Before you dive into your chapter, here's a 30-second map of Claude so you're not confused when we reference these later.
Claude Chat
This is the most powerful AI chatbot on the market — and it's where this entire course lives. Chat is where you have conversations with Claude: brainstorming, writing, analysis, research, problem-solving, strategy. You can connect it to your email, calendar, and files through connectors, which means it's not just answering questions — it's working with your real data. When you hear people say Claude changed how they work, this is what they're talking about. You'll spend most of your time here.
Claude Projects
A dedicated workspace where Claude always has your context loaded — your instructions, your uploaded files, your preferences. Every conversation inside a Project shares that knowledge. Think of it as giving Claude a specific job with permanent context. You'll set these up in your chapter.
Claude Code
You might hear about Claude Code — that's Claude's coding and developer tool. It runs in a terminal and is built for software engineers and technical work. We won't be using it in this bootcamp, and you don't need to learn it right now.
The simple version
Chat = your AI partner for everything (this is where the course lives). Projects = dedicated workspaces where Claude always knows your context. You'll use both constantly. Your chapter walks you through setting up each one for your specific job.
Done — setup complete
Claude Pro active, memory on, desktop app installed, global instructions set, response style configured. You know what Chat and Projects are. You're ready for your chapter.
You're ready
Setup done. Now go find your chapter, work through it in order, and by Sunday night you'll have something that makes Monday morning feel completely different. Let's get to work.