Fatchum 2.0

Fatchum 2.0 is a recipe recommendation app built to help users discover delicious recipes using the ingredients they already have on hand. It's a simple, smart, and efficient way to reduce food waste and find your next favorite meal.

Features

Mobile App (Kotlin)

• Search recipes by typing ingredients you have in your fridge.
• Get suggestions by food tags
• Firebase integration that supports Google Accounts

Backend (FastAPI + ML)

• Count Vectorizer + Cosine Similarity -> tag-based recommendations.
• TF-IDF + Cosine Similarity -> ingredient-based search engine.

Tech Stack

Mobile (Frontend)

• Kotlin + Andorid Studio
• Firebase (Auth, Firestore, Cloud Storage)
• Retrofit for API calls

**Backend

• FastAPI
• Scikit-learn (TF-IDF, CountVectorizer, Cosine Similarity)
• Uvicorn / Gunicorn for deployment
• Render for server hosting