Issue #6: Stop Downloading. Start Building

New here? If this landed in spam or promotions, drag it to your inbox and mark it "not junk" so you don't miss next week. Takes five seconds and it genuinely helps.

Here's the one I've been nervous to write.

For as long as I can remember, there was a door in my head marked "code," and it had somebody else's name on it. Not mine. Real developers. People who wear the right hoodie and understand what a bracket is for. I was a finance guy. A runner. A dad. I was allowed in a lot of rooms. Not that one.

So for years, when I needed something, I did what you do. I went to the app store. I typed in what I wanted. I scrolled through forty apps that did almost the thing. Wrong layout. Too many buttons. A monthly fee for the one feature I actually needed. And I picked the least-wrong one and I settled.

We all settle. We're so used to it we don't even feel it anymore.

Then a friend of mine, a colleague, built his own app. Not a company. Just a thing he wanted that didn't exist, so he made it. I watched him do it and something cracked open in my chest. A tiny, dumb, hopeful thought.

Wait. Could I do that?

I went home that night, opened Claude, and typed something like: "I want to build my own note-taking app. The ones I've tried don't work the way my brain works. Can you help me?"

I did not know what I was doing. I want to be really clear about that, because it's the whole point. No plan, no training, no clue. I just described what I wanted, in plain English, like I was texting the friend who knew how. And it started building.

That was the beginning. Hold onto that feeling, because it's the actual lesson this week.

Read that again. You don't have to hunt the app store for something close enough and settle. You describe exactly what you want, and it gets built. For you. To fit you.

That's a different world. And you already live in it. You just haven't tried the door yet.

Use case: Building a small, useful tool for yourself, right there in the chat

This is the doable-today version. You paste it into Claude, it builds you a little working tool right inside the chat, and you can actually use it. No download. No new account. Fill in the brackets and go.

Then here's the part almost everybody skips. You try it, and you talk back.

Watch it rebuild, right there, better. Do that two or three times and you've got a thing that fits you. You didn't find it. You made it.

The thing that just built you a tool inside the chat has a name. It's called an Artifact. Don't let the word do anything to you. An artifact is just a little app Claude builds you right there in the conversation, in its own window, that you can use and even share. No install. No setup. It's the difference between Claude telling you how to do something and Claude handing you the thing that does it.

Here's why this matters more than it sounds. Most people think of Claude as something that gives you words. Answers, emails, summaries. Words. An artifact is the moment it starts giving you tools instead. Once you've watched it do that, you stop looking at your everyday annoyances the same way.

Because here's where it goes. My note-taking app didn't stay a little window in a chat. It grew. To make it a real app that lives on the internet and remembers my stuff, I learned two more things by trial and error, with Claude guiding me the whole way. Vercel, which is just where your app lives online, so you can open it from your phone. And Supabase, the filing cabinet that remembers your data, so your notes are still there tomorrow.

I'm naming them so the words stop being scary, not because you need them today. Real talk: that's a later lesson, built over weeks, not an afternoon. Today's step is the artifact. Build the little thing in the chat first. That's the win this week. We'll climb toward the rest together.

Free plan. Browser. No download. Five minutes. Right now.

Open Claude the way you always do. Think of the single most boring, repetitive little calculation or list you deal with. Splitting a bill. A packing list you rebuild every trip. Tracking your glasses of water. Something small and real.

Paste the Prompt of the Week and fill in the brackets with that thing.

Then, when it builds it, use it. Actually poke the buttons. And then tell it the one thing you'd change. Bigger text. A different color. One more field. Watch it fix it.

That's the entire skill. You just described a thing that didn't exist, and now it exists, and it's yours. If your hands are a little shaky doing it, good. Mine were too.

I'll go first, like always.

It didn't stop at the note-taking app. Once that door was open, I couldn't stop walking through it.

I've shifted my training lately, away from the triathlon stuff I used to live for, toward getting genuinely strong. And I wanted to track my lifts properly. Not in my head, not in some app built for somebody else's body. So I built one. My own. I log every exercise and every weight, and here's the part that still makes me grin: the data goes into a database Claude can actually read. So it doesn't just store my workouts. It coaches me from them. It looks at what I lifted Monday and tells me what to reach for Thursday.

Then I built a little math app to help my kid practice. The ones I found were cluttered with cartoons and ads, and I wanted a clean one that did the exact drills he needed. So I made it.

I am not an engineer. I never wrote a line of any of this the way a programmer would. I described what I wanted, looked at what came back, said what was wrong, and it fixed it. Over and over, until it was right. (Even my 8-year-old has picked up enough of this to build his own stuff now. It's really not the wall you think it is.)

Here's the reframe I want to leave you with. For years I thought building software was something that happened to me, from the outside, made by strangers, take it or leave it. Now it's something I do. Not because I got smarter. Because I got curious enough to type the first sentence, and brave enough to look a little dumb for a while.

So here's your hand-off, and it's one thing.

Open Claude. Pick the smallest, dumbest, most annoying little task in your day. Paste the Prompt of the Week. Build the tool. Then tell it one thing to change, and watch it change it.

That's it. You'll feel the door open. I promise you will.

And listen, if you're the type who wants to run ahead, run. If you want to build real apps for yourself, put them online, get more sophisticated with the whole thing, you don't need to wait for me. Just talk to Claude. In chat, in co-work, in code, it honestly does not matter which one you're in. Tell it what you're trying to do and it will help you get there. That's the beauty of this. There's no wrong door.

And if that's not you, if you'd rather just hang here with me and take it a week at a time, that's perfect too. Stay right here. We'll get more advanced together, and I'll keep dropping you clues on how to evolve your skills as we go.

Next week we slow down. We've moved fast: chat, then co-work, then code, three rooms in a few weeks. And I've been walking you right past good stuff hiding inside each one. Skills. More on artifacts and what they can really do. Connectors, the thing that lets my workout app feed its data straight to Claude. We're going back through, slower this time, and I'll put those in your hands one at a time.

And somewhere in there, I still owe you the big one. Not what any of this does. Why it matters. Why I'm up at night writing this to you. That's the one I'm building toward.

Until then.

Stop overthinking it.

Just get started.

JGS

P.S. Hit reply and tell me the little tool you built, even if it's the dumbest tip calculator ever made. Especially if it is. Those are my favorite emails to get, and yours might be the one I show off next.