You’re going to build a real app. Not a pretend one — a program with a window and buttons that actually runs on this computer.
You won’t type the code. You’ll describe what you want to an AI, it writes the code, and then you test it, find what’s wrong, and tell it what to fix. Over and over until it’s good.
By the end of today you will have an app that works, that you made.
| What we’re doing | |
|---|---|
| 1 | Watch your instructor build an app in about 5 minutes |
| 2 | Learn what just happened |
| 3 | Everybody builds the same app: a Mad Libs story maker |
| 4 | Learn the three moves: Ask, Run, Fix |
| 5 | Make it yours — change it however you want |
| 6 | Try to break your own app (this is the fun part) |
| 7 | Show everybody what you made |
Open PowerShell. Type these lines one at a time, pressing Enter after each:
cd $HOME\Documents\Projects
mkdir <Name>
cd <Name>
mkdir madlibs
cd madlibs
claude
Put your name where it says <Name> — one word, like maya, so it works in the
commands. Your name folder is where all your projects will live. The first five
lines get you into today’s project folder, and that last one, claude, starts
the AI.
To quit the AI: hold Ctrl and press C.
To start it again: type claude.
To run your app (do this in PowerShell, not in the AI):
python story.py
Don’t put two commands on the same line with &&. It doesn’t work on these computers.
One line, press Enter, next line.
It’s a very, very fast guesser.
It has read an enormous amount of code, and it’s extremely good at guessing what code should come next. That’s what it does. It is not thinking about your project, and it does not know what you meant unless you said it.
Think of it like a genius intern on their first day. Unbelievably fast, knows every trick — and has no idea what you actually want. It will never tell you it’s confused. It’ll just confidently hand you something wrong.
This is the most important thing today: the AI can write code that runs perfectly and is still completely wrong. Your computer will never tell you the difference.
That’s your job. That’s why you’re here and not just watching a robot.
A good ask answers four questions:
| Question | Example |
|---|---|
| What kind of thing? | “a Python app with a window” |
| What does it do? | “asks for 5 words, then makes a funny story” |
| What does it look like? | “a text box for each word, and a big button” |
| Where does it go? | “save it as story.py” |
Here’s today’s first prompt. Notice it answers all four:
Build a Mad Libs generator in Python with a tkinter window.
Ask the user for 5 words: a noun, a verb, an adjective, a place, and a food.
Each one gets its own text box with a label.
When they click a "Make Story" button, show a funny story using their words.
Save it as story.py
See what’s not in there? Anything about how to write the code. That’s the AI’s job. Yours is deciding what gets built.
Ask for ONE thing at a time.
Starting a brand new app? Describe the whole thing.
Changing an app that already works? One change. Then run it. Then the next one.
Here’s why. If you ask for five changes and something breaks, you have no idea which one broke it. If you ask for one change and something breaks — well, it was that one. Obviously. You fix it in ten seconds.
This isn’t about being careful. It’s about being fast.
Ask small → Run it → Say exactly what went wrong → Repeat
That’s the whole class. Everything else is practice.
Three things can happen when you run your app:
Don’t say “it’s broken.” The AI has no idea what that means.
Say three things:
What I did: I clicked the Make Story button What I expected: a story should show up What actually happened: nothing happened, and there’s no error
That’s not an AI trick. That’s how you report a problem to any human being, ever. It’s a real skill and you’re learning it right now.
Near the end of class, everyone tries to break their own app on purpose. Try all of these:
Found something broken? Fix it. That’s your last change of the day.
Everybody’s app breaks. Everybody’s. Finding it is the win.
Try these in order:
| Word | What it means |
|---|---|
| Prompt | What you type to tell the AI what you want |
| Terminal / PowerShell | The black window where you type commands |
| Run it | Actually starting your app to see if it works |
| Bug | Something wrong with your program |
| tkinter | The Python tool that makes windows and buttons |
| Crash | Your program stops and shows an error |
At the end you’ll show what you made. Two things:
Number 2 is not optional! Everybody’s app broke today. Saying so out loud is the point.
Your app is a file on this computer. Save a copy to your Google Drive folder:
Ask an instructor if you can’t find your folder.
One last thing, and it’s honest: the AI wrote the code. You decided what to build, you found what was wrong, and you fixed it — that’s real, and it’s most of what the job actually is. But you don’t know how the code works yet.
If you want to get properly good at this, here’s the next step: open your own code and ask the AI
Explain what this part does, like I'm 12.
The people who can both steer the AI and read the code are going to be incredibly good at this.
This Vibe Coding Curriculum by Brian Bird, created in with AI assistance, is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.