AI App Builders
Gervigreindar-appsmiðir
A live app with their AI inside it.
What they'll learn
- Build a real web app Write the HTML, CSS and JavaScript behind a working app — not a toy, something that runs in a browser.
- Wire in real AI Call an AI model straight from their own code and turn its responses into a feature people can use.
- Design the AI feature Decide what the AI should and shouldn't do, and constrain it so it behaves reliably.
- Ship it live Deploy with Git to a real, shareable URL they can send to friends and family.
A deployed web app with a working, well-behaved AI feature — visited live by the showcase audience.
The 13-week journey
-
Sprint 1 — Your app shell
HTML/CSS structure and a clean interface; just enough JavaScript to make it interactive; deployed to a live URL in week 4. Ends with a working app shell online.
-
Sprint 2 — Add the AI
Calling a managed AI API from code; designing the feature (summarise, generate, answer); writing a system prompt and guardrails. Ends with the AI feature working live.
-
Sprint 3 — Make it trustworthy
Handling bad/empty AI output, a usage disclosure on the page, and polish; cost-awareness with capped keys. The honest-AI module lands here.
-
Week 13 — Showcase
Live URLs and QR cards; the audience uses each app's AI feature on their own phones.
What we cover
Every topic, unit by unit — so you know exactly what your child builds and learns.
Your app shell
- HTML structure and clean CSS
- Just enough JavaScript to be interactive
- Deploying to a live URL in week 4
- A working app shell online
Add the AI
- Calling a managed AI API from code
- Designing the feature: summarise, generate, answer
- Writing a system prompt and guardrails
- The AI feature working live
Make it trustworthy
- Handling bad or empty AI output
- A clear AI-use disclosure on the page
- Cost-awareness with capped keys
- Polish and accessibility basics
Ship & showcase
- Git and deploy workflow
- Final testing on a phone
- Live URLs and QR cards for the audience
A live app on the big screen with a working AI feature and a QR code the audience opens on their phones.
The course where AI stops being something you chat with and becomes something you build into a real product. By week 13 each teen has a live web app with their own AI feature inside — shareable, on their phone, theirs.
The hooks
Kid hook: “I built an app with my own AI in it — here’s the link.” Parent hook: “A live, shareable AI product with their name on it — real building, not just prompting.”
Who it’s for
No prerequisites; pairs naturally after AI Makers. Thrives: builders, the project-driven, and teens motivated by a real audience.
Outcomes — by the end, students can
Build and style a small web app; make it interactive with JavaScript; call an AI model from their own code and shape its behaviour; handle failure cases and disclose AI use; version and deploy with Git to a live URL.
Tools & compliance
VS Code, Git and a static host (13+ accounts with signed parental consent), school-managed AI API keys with spending caps, content guidelines for anything published under the school’s umbrella.
Where this course fits
Follows AI Makers; leads into LLM & Agent Builders at 16–18.
Parent questions
Do they need to know how to code first?
No — we build the web basics in the course; AI Makers helps but isn't required. A placement chat finds the right depth.
What does the AI feature cost us?
Nothing — API access runs on school-managed, capped keys.
Who controls what gets published?
Deploys run through school accounts and content guidelines; we review before anything is public.
The first lesson is a free trial.
Book a no-commitment trial — pay nothing if it's not a fit.
Book a free trial