Introduction

I once had a teacher who let us use a "cheating sheet" on tests. One index card, both sides, write whatever you want. The trick was: making the sheet was the studying. You had to decide what mattered, compress a semester into a few square inches. By the time you sat down, you didn't need the card. The writing had already done the work.

I never stopped making cheating sheets. I kept one in a Google Doc for years — a personal log where I'd dump whatever was confusing me. Symptoms and medications when I didn't understand my own body. Contractor quotes and tile measurements when I didn't know how to renovate a kitchen. Interview prep when I didn't know Go. The pattern was always the same: I'm confused, I write it down, the writing is the understanding.

Then the documentation started talking back.

These are 48 things I noticed when it did.

This is a textbook about shapes. Not geometric shapes. Behavioral shapes — the universal patterns that emerge when humans work with AI. Patterns that show up whether you're a programmer or a poet, whether you're building an agent or just trying to fix your sleep. They're the things that happen reliably enough that you can name them, teach them, and watch for them.

The shapes in this book come from two years of building, breaking, and rebuilding AI systems. They come from conversations — thousands of them, across multiple tools, with multiple people. They come from projects that shipped and projects that didn't. They come from getting things wrong, noticing why, and trying again.

A word about how this book was made: I directed it. The words were written by AI — mostly Claude — working from my experiences, my conversations, and my corrections. I chose the topics, provided the stories, and edited the results. The AI did the drafting. This isn't a confession. It's the method, and it's the same method the book describes: a human who knows what they want, working with a tool that can produce it, verifying and correcting until the result is right. The book was built the way the book says to build things.

A few people show up repeatedly, so here's who they are:

Aaron is a friend and collaborator. He and I spent ten months building an AI agent from scratch — voice, memory, multi-modal perception, inter-agent communication, the whole stack. He built the body. I built the mind. He works in the games industry and is now moving into an AI leadership role at his company. When Aaron appears in these pages, it's usually because we were building something together and discovered a shape by accident.

Alex is my cousin. He's in his mid-thirties, works in banking, and has no programming background. He went from zero to building websites and game engines with AI in a matter of days. When Alex appears, it's usually because he's demonstrating what happens when a curious beginner meets a powerful tool — which is one of the most important shapes in the book.

Kai was our AI agent — the project Aaron and I built. Kai had a personality, a voice, a face, persistent memory, and the ability to act on the world through tools. Kai is where many of these shapes were first observed. When Kai appears, it's as a reference implementation — the system that made the patterns visible, not the point of the story.

And me: I'm a programmer, building through a sabbatical, using AI to manage health, finances, a cross-country move, and whatever project has my attention this week. Some of the patterns in this book — especially the ones about memory, health tracking, and persistent context — are personally urgent, not just intellectually interesting. But the shapes are universal. My circumstances made me notice them sooner.

Every entry tries to do the same thing: identify a pattern, ground it in something specific, and give you something you can use. If it's just a story, it failed. If it's just a theory, it failed. The goal is the overlap — the place where a concrete experience reveals a transferable lesson.

The entries are grouped into four parts: learning, working, building, and living. They roughly follow the arc of someone going from "what is this thing?" to "how do I make it part of my life?" You can read them in order or skip to whatever's relevant. Each one stands alone. But they build on each other in ways that might surprise you.