Unit 9Build something real

The big idea

You don't have to be qualified to start building. AI has removed most of that barrier. The goal isn't something perfect. The goal is one thing that works, and the habit of building.

Why it matters

School teaches that you need permission and qualifications before you do things. You don't. A weekend and two chat windows are enough to build the first pieces of an idea. The people who start now will be years ahead.

Pick one path

1. Personal portfolio website. Use a free AI website builder. Everyone should have one.
2. A simple image classifier. Use a free tool that lets you train a model to recognize a few objects from photos you take.
3. Your own project. A first version of the idea behind your three questions.

Do this

1. Plan before you build. Take your three questions, the research you've done, and sketch the big blocks of your idea. Then break each block into small tasks.
2. Pick one feature. Just one. "A page that shows my three best projects." "A model that tells apples from oranges." "A form that collects one piece of information."
3. Build it with the loop. Plan in one window, build in another, review, fix (Unit 6).
4. If you're stuck, tell the AI. Literally type: "I don't know how to go from here." Then let it ask you questions.
5. Test it on something new. Try an input the build has never seen: a different background, a different user, a weird question.
6. Let someone else use it. Watch without helping. Where did they get confused?
7. Fix the biggest problem you saw.
8. Write down one limitation. Every build has one. Naming it is part of the work.

Your portfolio website

Decide what goes on it before you open the builder. Good blocks:

• Who you are, in two sentences
• Your three questions
• What you've built, with a screenshot each
• What you're reading or learning
• How to reach you

Show the builder examples of sites you like. Tell it what you like about each. If you can, buy your own name as a domain. One day you'll build something people want to look up, and it should be under your name.

Purpose before product

A student once asked me whether he should build an app. My answer: anyone can build an app now. The question is why. Why is it important? Who does it help? Why is that a problem worth solving? How does it make money? Answering those is far more important than building the app.

When a big ticketing company offered me their platform for free, I turned it down and built my own, because I knew exactly what theirs couldn't do. The first version took a team a month. Rebuilding it later with AI took a week.

The two-window method for apps

1. Window 1 plans the app: the features, the screens, what it must do.
2. Window 2 builds it.
3. Window 1 reviews what window 2 built and gives feedback.
4. Window 2 fixes it.
5. Repeat, then launch it with a free website builder or host.

Questions to ask yourself

• What's the one feature that would make this useful to one real person?
• What happened when someone else tried it?
• What's the limitation I'd admit to first?

In my words

"The goal is not to look for something perfect. The goal is just to get something and get used to building."

"The most important thing is not learning what other people do. It's learning how to do it yourself."

"There's no small step for building. It's just getting stuff done and moving to the next step, and the next step."

Worksheet

Use Worksheet 9: Build sprint card in the PDF kit.