The story behind the island

I thought I'd add a bit here about me.

This game is a birthday gift for the lovely Sophie.

I'm the creator of this game. Every detail, every system, every quest and every little in-joke was carefully crafted to make sure the whole thing flowed well and felt balanced.

I didn't write a single line of code for it, though.

I've seen a lot of discussion online about the rise of AI, and plenty of arguments about AI generated slop and the damage it can do to artists. I understand those concerns, but I wanted this project to show that it isn't always that simple.

I had the vision for this game, but I didn't have the technical know how to build it on my own. This is only the second game I've ever made, and it's by far the biggest project I've taken on.

The game was built entirely with Anthropic's Claude Fable 5 and Opus 4.8 models. The intro video and loading screen were created using Gemini Flash 2.5. For me, this game is proof that AI can be a creative tool rather than a replacement for creativity. Without these tools, this game simply would not exist.

Anyway, onto the fun bit.

The music

This project threw plenty of challenges at me. The first big one was the music. I knew I wanted to build a synth engine and generate the soundtrack inside the game rather than creating songs in Suno, which I did experiment with for a while. I wanted something that felt relaxing, looped naturally and didn't become irritating after twenty minutes.

At first I kept setting a key, choosing a BPM and letting it generate arpeggios. It just wasn't right. Eventually I ended up writing a simple melodic line myself, then prompting the synth engine with exactly which notes to play, how long to hold them, which waveforms to use and how many layers to stack together. There were countless iterations before it sounded right to my ears. Music is subjective though, so if you hate it… sorry!

Pacing & economy

Another huge challenge was balancing the pacing and the economy. Making sure the game flowed properly, placing gates on quests so players couldn't race through the main story, and making sure everything felt rewarding without turning into a grind meant endless tweaking. Every little change had a knock on effect somewhere else.

Then there's me

I'm my own worst enemy.

When I started, I had a proper plan. First I'd map the island, then create the characters, then build the mechanics like brewing, festivals and the shop. After that would come the quests, then polishing, then the build itself. I'd even written a full play testing checklist.

It lasted about five minutes.

I'd get halfway through a playthrough and suddenly think of an entirely new mechanic. That's how the garden plots happened. Then the greenhouse. Then a complete overhaul of the object placement system. Then I rewrote the final three acts.

If I'd stuck to the original plan, the game would probably have been finished in half the time. It also wouldn't have been anywhere near as good.

One of my favourite moments is simply walking out of the brewery at dusk and seeing the lighting on Sophie's face.

Chef's kiss.

Lighting and graphical improvements became another rabbit hole that took far longer than I'd ever planned, but I'm so glad I spent the extra time on them.

Overall, I'm incredibly happy with how this turned out. It definitely won't be my last attempt at making games with AI.