MzeeChakula - AI-Powered Nutrition Assistant 🇺🇬

MzeeChakula is an intelligent nutrition assistant designed specifically for the elderly in Uganda. It leverages advanced AI to provide personalized dietary advice, meal planning, and health monitoring, all accessible through a user-friendly interface with support for local languages and voice interaction.

Live Demo

• Frontend (User Interface): https://mzeechakulaaissistant.vercel.app
• Backend (API Docs): https://mzeechakula-user-interface-backend.onrender.com/docs

System Architecture

+-------------+          +------------------+          +------------------+
|   User      | -------> | Frontend (Vue 3) | -------> | Backend (FastAPI)|
+-------------+          +------------------+          +------------------+
                                                                |
                                                                |
                                       +------------------------+------------------------+
                                       |                        |                        |
                                       v                        v                        v
                              +----------------+       +----------------+       +----------------+
                              |    Supabase    |       |    ChromaDB    |       |   AI Services  |
                              |  (PostgreSQL)  |       | (Vector Store) |       | (OpenAI/Groq,  |
                              +----------------+       +----------------+       |  Sunbird, HF)  |
                                                                                +----------------+

Key Features

• AI Nutritionist Chat: Interactive chat powered by RAG (Retrieval-Augmented Generation) to answer questions about nutrition, specifically tailored to Ugandan local foods and dietary needs.
• Smart Meal Planning: Generate personalized weekly meal plans based on health goals, dietary restrictions, and available ingredients. Export plans to PDF.
• Voice & Language Support:
  • Voice-to-text and text-to-speech capabilities for accessibility.
  • Real-time translation for local Ugandan languages (Luganda, Acholi, Ateso, Lugbara, Runyankole) powered by Sunbird AI.
• Food Analysis: Advanced nutrition analysis using an ensemble of machine learning models (XGBoost, Hugging Face embeddings) to estimate calories and nutritional content.
• Document Analysis: Upload medical or dietary documents (PDF) for the AI to analyze and incorporate into its advice.

Tech Stack

Frontend

• Framework: Vue 3 (Composition API)
• Build Tool: Vite
• Styling: TailwindCSS + Custom CSS
• State Management: Pinia
• Routing: Vue Router
• HTTP Client: Axios

Backend

• Framework: FastAPI (Python)
• Database: Supabase (PostgreSQL)
• Vector Store: ChromaDB (for RAG)
• AI/ML:
  • LangChain (Orchestration)
  • Hugging Face (Embeddings & Ensemble Model)
  • OpenAI / Groq (LLM)
  • Sunbird AI (Translation)
• Deployment: Docker

Local Development Setup

Prerequisites

• Node.js (v18+)
• Python (v3.10+)
• Docker (optional, for containerized run)

1. Backend Setup

cd backend

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Set up environment variables
cp .env.example .env
# Edit .env with your API keys (OpenAI, Supabase, Sunbird AI, etc.)

# Run the server
uvicorn api.main:app --reload

The API will be available at http://localhost:8000.

2. Frontend Setup

cd frontend

# Install dependencies
npm install

# Set up environment variables
cp .env.example .env.local
# Ensure VITE_API_BASE_URL=http://localhost:8000

# Run the development server
npm run dev

The application will be available at http://localhost:5173.

Docker Deployment

The backend is fully containerized.

# Build and run with Docker Compose
docker-compose up --build

Testing

The project includes comprehensive test suites for both frontend and backend.

• Frontend: npm test (Vitest)
• Backend: pytest

License

This project is licensed under the MIT License.