Niblet
A native iOS virtual pet fish with personality. Niblet chases lasers, begs for food, gets bored if you ignore it, and has opinions about the time of day. It doesn't do anything useful. That's kind of the whole point.
The question
Remember those old desktop pets from the 90s, the ones that waddled across your screen and made you smile for no reason at all?
What if you brought that to your phone, but made it feel genuinely alive? Not a productivity tool, not solving any real problem. Just a little fish that makes you happy.
The itch
Screens stopped being fun
Most of the time I spend on them is productive, focused, serious. Which is fine. But I kept coming back to one question: what if the silliest possible thing I could build was also the most joyful? No AI, no automation, no productivity gains. Just a fish named Niblet who wants your attention.
What happened
A weekend project that grew a personality
Niblet started as a weekend project: an iOS app with a fish that responds to touch. Then the personality crept in. Niblet begs for food, chases laser pointers with manic energy, and gets genuinely bored if you ignore it. People started sending screenshots to friends saying "look at this ridiculous fish." That felt like enough.
Niblet mid-chase, tail going, absolutely certain the laser dot is catchable this time.
Niblet is needy, easily distracted, and unapologetically useless.
That's the whole point. Free on the App Store. No accounts, no nonsense. Just a fish.
Get Niblet on the App Store →› A note on how it's built
Built entirely with Claude Code over a weekend. Swift and SwiftUI for the animations, particle effects and custom lighting for atmosphere, a simple state machine for moods and behaviours, and touch tracking with gesture detection for the interactions. The fish's moods shift based on time of day, which I didn't plan but which felt right once it was in.
- Swift and SwiftUI animations with particle effects and custom lighting
- Simple state machine for moods and behaviours
- Touch tracking and gesture detection for laser and feeding interactions
- Built entirely with Claude Code as a weekend project