It was acceptable in the 80s

#deved #learngoogleaistudio #ai #gemini

Education Track: Build Apps with Google AI Studio

This post is my submission for DEV Education Track: Build Apps with Google AI Studio.

What I Built

For the Google AI Studio Education Track, I wanted to build something a bit different. So I dusted off my inner 80s child, loaded up my imaginary tape deck, and created…

🎨 A ZX Spectrum–style Loading Screen Generator
Built entirely using Google AI Studio, Gemini, and Imagen.

Think chunky pixels, screaming BRIGHT colours, and enough nostalgia to make your old rubber-key Speccy shed a tear.

Demo

So the application is simple, you provide it a prompt and it will generate a ZX Spectrum loading screen based on your prompt.

The app is deployed here, I have put a loose limit on usage as I did give in a sign up for the billing, but with a very low budget as I don't actually expect people to use it.

My Experience

Honestly?
Google AI Studio made this absurdly easy.

🚀 Fast

The build was really quick, I mean, like, really, like as fast as a Knight Industries car.

Within a minute, I had:

A full React + TypeScript project
API integration
A UI
Working image generation

It felt like watching a Speccy loading screen but… much faster and without the shrieking sound effects.

🧠 Iteration was Weekend At Bernie's dead simple

After the first version, I asked the assistant to make a few small changes or additions, and it was very smooth and still very quick.

🔄 GitHub Integration

The link between Google AI Studio and GitHub was so smooth it felt like cheating.

A couple of clicks and the repository was syncing perfectly.

☁️ Deployment

Google Cloud Run is great, but for familiarity’s sake, I deployed using Vercel, which worked flawlessly once I added the environment variables.

If you'd told teenage me — sitting in front of a 48K Spectrum — that one day I’d deploy an app with a single click instead of faffing about with LOAD ""… I’d have assumed you’d been inhaling too many cassette fumes.