Doodytoon
Built at Meta Global AI Developer Hackathon · Oct 26, 2026 · Remote

DoodyToon is an interactive storytelling studio that turns a hand-drawn character into a participant in its own story. Today, creators can draw or import a scene, then use text or voice to tell the character what should happen next. The character responds with a new story event, emotional state, and three possible branches. Each decision is added to a visible story path, so users can see how their choices change the narrative. The problem is that most creative AI tools stop at generating a single image or video. They help make the first frame, but not the shared authorship that comes after it. DoodyToon explores a different workflow: the creator supplies the visual idea, and the AI helps the character answer, react, and create the next frame of the story. The submitted prototype is a working React application with an HTML5 Canvas drawing studio, image-scene upload, browser voice interaction, server-side story generation, a deterministic demo fallback, and a branching story engine. It is designed as a practical, testable slice rather than a finished consumer product.