Capstone: Build and Deploy an Original Idea
Module 5 · Ship It

Capstone: Build and Deploy an Original Idea

Learning by doing, one last time: build something of your own choosing from scratch, then make it reachable by other people.

Stuck on what to build? Here are six real projects, each with a paste-ready prompt. Use one as-is, or change a few words to make it yours. Every one of these is finishable in a single sitting.

☕ Tip Splitter

“Build a tip calculator: inputs for the bill amount, a tip percentage, and the number of people. Show the tip, the total, and the amount each person owes, updating live as I type. Clean and mobile-friendly, warm orange theme.”

What you’ll see: three inputs and a running breakdown you could actually pull up at dinner.

⏳ Countdown to a Big Day

“Build a full-screen countdown page to a date I choose, showing the days, hours, and minutes remaining, with a celebratory message when it hits zero. Bold and colorful.”

What you’ll see: a live ticking countdown to your next trip, birthday, or deadline.

✅ Habit Tracker

“Build a daily habit tracker: I can add habits, check each one off for today, and see a streak count for each. Save my habits so they’re still there when I reload the page.”

What you’ll see: checkboxes and streak counters that remember your progress between visits.

🎲 “What Should I Eat?” Decider

“Build a decision-maker: I type in a list of food options, click a button, and it randomly picks one for me with a fun little animation.”

What you’ll see: an editable list and a satisfying pick-for-me button.

🔗 Personal Link Page

“Build a simple personal link page: my name, a short bio, a photo placeholder, and buttons linking to my social profiles. Modern, minimal, dark theme.”

What you’ll see: a tidy link-in-bio page you could actually share.

📇 Recipe Card Collection

“Build a recipe collection: I can add recipes with a title, ingredients, and steps, and browse them as cards. Save them between visits.”

What you’ll see: a grid of recipe cards that persists each time you come back.


🛠️ Learning by Doing

This is the whole course, one more time, at full size. No new instructions to memorize — just the four skills you already have, applied to something you pick.

💻 Code Lab: Step 1 — Pick a Real Idea

Pick one project from the gallery above, tweak one of them, or bring an idea entirely your own. The only rules: small enough to finish today, and genuinely yours.

Choose the one that makes you want to open it right now.

💻 Code Lab: Step 2 — Describe, Then Build (Module 1)

Write your first prompt. Be specific about what it does and how it should feel — the same way you did in Module 1’s first build. Run it.

💻 Code Lab: Step 3 — Loop Through It (Module 2)

Run at least two real rounds of describe → test → refine. Fix something that bugs you, then add one thing that wasn’t there before.

💻 Code Lab: Step 4 — Expect and Handle a Break (Module 3)

If something breaks along the way — and with a new idea, something likely will — report it the way you practiced: what you did, what happened, what you expected instead. Don’t dread it. You already know this part.

💻 Code Lab: Step 5 — Run the Stranger Test (Module 4)

Before you call it done, run the 5-second test, the thumb test, and the silence test on it. Fix the single worst offender.

💻 Code Lab: Step 6 — Ship It

Every vibe coding tool has some version of a publish or deploy button — look for language like “Publish,” “Deploy,” “Share,” or “Get a link.” Use it.

If your tool doesn’t have a one-click option, ask it directly:

“How do I host this so other people can open it with a link?”

Once you have a link, open it in a new browser tab or on your phone. If it loads there, it’s genuinely live — not just visible in your own preview.


🔍 What You Just Did

You just went from a blank idea to a live, shareable app, using nothing but the same four moves you first practiced on a simple about-me page: describe, refine, debug, evaluate. The idea got bigger. The process didn’t change at all.

⚠️ Common Pitfall: 'Finished' Is a Choice, Not a State

There’s always one more thing you could refine. That’s true of every piece of software ever shipped, including ones built by large professional teams. Waiting for “no more improvements come to mind” means never shipping.

Pick a reasonable stopping point — the stranger test in Step 5 is a good one — and ship at that point. You can always run another loop later; you can’t get feedback from real users on something that’s still sitting in a private preview.


⏭️ Up Next

One more lecture — what this course was actually preparing you for, and where to go from here.