DON'T BUY
THIS BULLSHIT
AI PLAYBOOK.

Every AI course has this page. So here's ours.

Still here? Fine. Don't say I didn't warn you.

The intro I threw away

There was supposed to be an intro here. I actually wrote it. Four pages on what AI is, how a language model works, and why this is the biggest thing since the internet.

Then I read it back and deleted it.

You already know what AI is. You want to save time. So do I. Making you read four pages of stuff you already know would be a strange way to start.

If you're looking for "What is ChatGPT?", it's free in roughly three million places. This book starts where those end.

Here's the deal. Everything in here I've built, used and timed myself. These systems don't just save me 40 hours a week. They fit four weeks of work into two days. I know how that sounds. So instead of asking you to trust me, I show you the screen, the time before and the time after. I also show you where the AI screws up. It does, every day, often with total confidence.

Don't buy this book if you enjoy explaining yourself to ChatGPT every morning. Some people find it relaxing.

Seriously. Don't buy this book if you…

The staircase

Everything you can do with AI sits on one of five steps. Most people are on the first one and think that's the whole staircase.

You're probably on step 1 or 2. This book takes you to step 4 or 5.

What's inside. Roughly.

Appendix: the vault. You'll know it when you get there.

Your Second Brain

How to build an AI assistant that actually knows you

Every time you open a new chat, you meet a stranger. Brilliant, sure. But it doesn't know your name, what you do, or that you hate texts that open with "In today's fast-paced world". So you explain. Again. And again.

It's like hiring an assistant with amnesia. Every single morning.

By the end of this chapter you'll have an assistant that knows you. It takes one evening. You'll notice it tomorrow.

Three levels of knowing you

1. The chat. A stranger every time. This is where almost everyone is.
2. The project. You write a page about yourself, and every chat in that project starts from it. Most people never get this far. This is where you start pulling away from the pack.
3. The second brain. It doesn't just read instructions. It writes them. It remembers, reflects, takes your feedback and rewrites its own rules. This is where we're going.

What a second brain actually is

It's not an app. It's an AI agent that lives in a folder on your computer. Claude Code, OpenAI's Codex, or whatever is best the day you read this. The brand matters less than you'd think. The setup is what matters.

The folder is its memory. Plain text files. You can open them, read them, change them, delete them. Nothing hidden in someone else's cloud settings.

The loop that makes it smarter every week

Here's what one of those lessons looks like. One fact per file, with the reason, so it can apply it with judgement instead of like a robot:

When it works like this

• You stop explaining. Monday's correction is Tuesday's rule.
• It picks up where you left off. Every session starts by reading the handover. No "so, where were we?"
• It notices things. At the end of the day it looks back, and flags the email you've ignored for a week before it becomes a problem.
• It lives where you are. A message on your phone, not another browser tab you forget to open.

Mine has a name. Giving it one felt silly for about a day.

Tonight: start at level 2

You'll build the full thing later in the book. Tonight, get the first page right, because it's the page your second brain will one day keep up to date by itself. Paste this into a project in ChatGPT or Claude:

Where it screws up

• It remembers the wrong things. Sometimes a fact changes or turns out to be wrong. A good second brain cleans up after itself: it updates the file, or deletes it. Your only job is to say "that's not true anymore".
• It can do things, not just say things. Give it hard rules for anything that leaves your computer. Nothing sent, posted, paid or deleted without your explicit yes.
• The tool will change. The folder won't. Your memory is plain text. Take it with you to whatever is best next year.

Never Do It Twice

How to turn your repetitive tasks into reusable AI templates

You have tasks you could do in your sleep. That's the problem. You're awake for them.

The weekly update. The "thanks, here's what happens next" reply. The cover letter you rewrite for every application. Fifteen minutes each. Nobody notices fifteen minutes. Add them up over a year and you've lost a holiday.

The third time you do something, it becomes a template. No exceptions.

A template that doesn't suck has four parts

1. Job. What it does and for whom. One sentence.
2. Slots. Marked places for the stuff that changes each time.
3. Shape. Exactly what comes out: length, structure, tone.
4. Check. What it verifies before answering. Everyone skips this part. That's why everyone's output reads like a brochure.

"Write a cover letter" is not a template. "Match my three strongest points to three things the ad asks for, max 200 words, and flag anything I can't honestly back up" is.

Two ways it goes wrong. Thin input gives you generic output, so the slots are where the quality lives. And not everything should be a template: an apology, a hard conversation, anything where the other person needs to feel you wrote it. Write those yourself.

Next chapter, we chain templates together. That's where it stops being a shortcut and becomes a machine.

Don't buy this book if you secretly enjoy rewriting the same email every Monday. It's a valid hobby.

Chaos In, Done Out

How to build your first AI workflow, step by step

A template does one thing. A workflow does the whole job.

Time doesn't leak in the doing. It leaks in the deciding. What first? Who am I late replying to? What did I forget? AI is very good at deciding, as long as it knows what matters to you. You sorted that in chapter 1.

So: talk for five minutes. Get a plan back.

Ramble into a voice memo. Everything on your plate, no order. Run this in your project from chapter 1. Read the result like a boss, not an employee. Send the drafts before bed.

See what DROP does? It uses TOP 3 to decide what to cut. That's chaining, and every workflow in this book works the same way. First time, it'll be too polite to drop anything. Tell it to be ruthless. It will be. Maybe a little too ruthless.

Every evening at 22:03 a summary of my day shows up on my phone. It's the most reliable relationship I have.

Stop Giving Orders

Everything so far, you drive. You paste, you ask, you remember to do it again next week. That's already more than most people ever do.

But the real shift is this: you stop giving instructions. You start getting asked.

Questions. Suggestions. "Should I go ahead?" Your job shrinks to the one thing only you can do: deciding. Yes. No. Not now.

It's the difference between owning a hammer and having a colleague who already knows which nail is loose.

Don't buy this book if you like being the one who remembers everything. Someone has to. It just doesn't have to be you.

Why a Second Brain

Here's what mine does in a normal week. Not because I tell it to every time. Because it knows me well enough to see what needs doing, and asks.

• Keeps track of every task, so I don't have to
• Remembers what I promised, and to whom
• Switches topics mid-sentence without losing the thread
• Answers my email. It drafts, I say yes
• Plans my training around my calendar
• Keeps score on my hobbies
• Makes the images
• Builds the websites
• Runs the ads, and tells me which one is burning money
• Runs an online store
• And a bunch more stuff.

Here's how to set it up: