Raseed AI - Expense Parser & Analyst

A complete AI-powered expense parsing and analysis system with a Flask backend and React frontend. The system can parse natural language expense descriptions, automatically calculate shared expense splits, and provide AI-powered financial insights.

Architecture

• Backend: Flask API with AI-powered expense parsing using Vertex AI (Gemini)
• Frontend: React application with modern UI components
• Database: SQLite for expense storage
• AI Integration: Google Vertex AI for natural language processing

Features

• 🤖 AI Expense Parsing: Describe expenses in natural language
• 👥 Smart Shared Expenses: Automatically detects and calculates splits
• 📊 Financial Analytics: AI-powered insights about spending patterns
• 💬 Chat Assistant: Natural language queries about your expenses
• 📱 Modern UI: Clean, responsive interface

Quick Start

Prerequisites

• Python 3.8+
• Node.js 16+
• Google Cloud Platform account with Vertex AI enabled
• GCP authentication set up

1. GCP Setup

Enable Vertex AI API

gcloud services enable aiplatform.googleapis.com

Authenticate (Choose one method)

# Method 1: Application Default Credentials (Recommended for development)
gcloud auth application-default login

# Method 2: Service Account Key
export GOOGLE_APPLICATION_CREDENTIALS="/path/to/your/service-account-key.json"

2. Backend Setup

# Install Python dependencies
pip install -r requirements.txt

# Set your GCP configuration
export GCP_PROJECT_ID="your-gcp-project-id"
export GCP_LOCATION="us-central1"  # Optional, defaults to us-central1

# Start the Flask backend
python app.py

The backend will run on http://127.0.0.1:5002

3. Frontend Setup

# Install Node dependencies
npm install

# Start the React frontend
npm run dev

The frontend will run on http://localhost:5173

API Integration

The frontend now uses all backend APIs:

• Expense Parsing: Natural language → Vertex AI processing → Structured data
• Data Persistence: All expenses stored in database
• AI Insights: Chat queries → Backend analysis → Personalized responses
• Authentication: Bearer token system (default: 'local-user')

Usage Examples

Adding Expenses

"Yesterday I went to Big Bazaar with Priya and Rohit. We bought groceries worth 2500 rupees including rice, dal, vegetables, and snacks. We split the bill equally."

Development

Use the convenient development script:

./start-dev.sh

This will set up environment variables, install dependencies, and start the backend.