Sonoro Studio is an independent studio. I design and ship desktop tools, iOS games and short animations, working with frontier models every day — from code to visuals to voice.
Small, focused tools and experiences built end-to-end by one person with AI coding agents, image models, and speech synthesis.
A desktop companion that watches your AI usage.
A frog that changes mood with your token usage, cost alerts and agent quota windows across Codex, Claude and Gemini.
Merge your way through 4,000 years of Chinese history.
A 5×5 2048 where tiles evolve 夏 → 商 → 周 … → 中国, with a chronicle you unlock as dynasties rise and merge.
Tactile puzzle games for the phone.
Draw one continuous line through every lit dot; plus a trilogy of grid puzzles (ColorLink, TriMerge, PixelLogic).
Sophie’s World, retold as crayon doodle shorts.
One-minute vertical episodes, one big question each — blending hand-drawn warmth with programmable HTML canvas and SVG motion.
Public-domain museum paintings, each paired with a short first-person or curatorial narration generated with Gemini TTS.
“He called the cypress as beautiful in line and proportion as an Egyptian obelisk.”
「波のくぼみに、富士がひとつ。」 Narrated in Japanese.
“Nothing happens here — and that stillness is the whole story.”
All artworks are public domain. Voices are AI-generated.
AI is part of every stage of my studio workflow — while scope, architecture, and final quality stay human-owned.
One page, three things it won’t do. Keeping scope tight makes solo shipping sustainable.
Frontier models draft UI options, layout explorations, and storyboards before I refine the details.
Implemented with AI coding agents across Swift, Rust, and TypeScript — reviewed line by line.
Tested on real devices against real services, APIs, and edge cases before any release.
Hi, I'm sugarfaye. I build small things that feel good to use — desktop utilities, tactile mobile puzzles, and short educational animations. Sonoro Studio is my personal workshop for exploring what one developer can craft when pairing careful product taste with frontier AI models.