muscodes
← Projects
2024MobileAI

AI Chatbot

A cross-platform AI app that puts conversation, image generation, and voice into one place.


Problem#

Most AI apps do one thing. I wanted a single Flutter app where a conversation could flow naturally into generating an image, dictating by voice, or picking up exactly where you left off on another device.

Constraints#

  • One codebase, every platform: Android, iOS, web and desktop from the same Flutter project.
  • AI features behind a clean boundary so a provider can be swapped without touching the UI.
  • History that survives reinstalls and syncs across devices.

Build#

Conversations run on Google's Gemini API, and image generation is provider agnostic behind one interface, wired to DALL-E, Hugging Face, and Stability so none of them is load-bearing. Speech-to-text and text-to-speech make it usable hands-free. Firebase handles auth with Google Sign-In, and Firestore stores chat history so it follows you across devices, with offline use and cloud sync. In-app purchases gate the heavier features.

Result#

A genuinely multi-modal assistant from a single codebase. It is the project I keep coming back to, and the one I am preparing to publish on the Play Store.

Stack

FlutterDartGemini APIFirebaseFirestore