Welcome to your Expo app 👋

This is an Expo project created with create-expo-app.

Get started

1. Install dependencies

   npm install

2. Start the app

    npx expo start

In the output, you'll find options to open the app in a

• development build
• Android emulator
• iOS simulator
• Expo Go, a limited sandbox for trying out app development with Expo

You can start developing by editing the files inside the app directory. This project uses file-based routing.

Environment Variables

The app uses environment variables for API tokens and other sensitive information. Create a .env file in the root directory of the project and add the following variables:

API_TOKEN=your-api-token
USERNAME=your-username
OPENAI_API_KEY=your-openai-api-key

AI Providers

The app supports three AI providers:

• Ollama
• ChatGPT

You can switch between providers by setting the CURRENT_PROVIDER variable in the constants.js file.

Implementation Details

Architecture Overview

This is an Expo-based mobile application for expense management that integrates with Kimai timesheet system and AI-powered image analysis for automated expense entry.

Tech Stack

• Frontend Framework: Expo (React Native)
• Language: JavaScript
• AI Integration: Multi-provider support (Ollama, ChatGPT, Claude)
• External APIs: Kimai timesheet API
• File-based Routing: Expo Router

Key Features

1. Kimai Integration: Direct integration with Kimai timesheet system for expense tracking
2. AI-Powered OCR: Automatically extracts expense details from receipt images
3. Multi-AI Provider Support: Flexible configuration to use different AI services
4. Expense Management: Track expenses with categories, dates, and amounts

How It Works

1. Expense Image Analysis

The AI service processes receipt/invoice images through the following flow:

• User captures or uploads an expense receipt image
• Image is converted to base64 format
• AI service sends the image to configured provider (Ollama/ChatGPT/Claude)
• AI extracts structured data: date, category, description, and amount
• Extracted data is returned in JSON format for further processing

2. Kimai API Integration

The application connects to Kimai timesheet system to:

• Fetch expense categories
• Submit expense entries
• Retrieve customer and project data
• Manage activities and timesheets

Authentication is handled via custom headers:

• X-AUTH-USER: Username
• X-AUTH-TOKEN: API token

3. AI Provider Configuration

The app supports three AI providers with a unified interface:

Ollama (Local/Self-hosted):

• Uses ngrok tunnel for remote access
• Model: llama3.2-vision

ChatGPT:

• OpenAI API integration
• Model: gpt-4-turbo
• Vision capabilities for image analysis

Data Flow

1. Image Capture → Base64 Encoding
2. AI Analysis → Structured JSON extraction
3. Data Validation → Category mapping and formatting
4. Kimai Submission → Expense entry creation

Configuration

Switch AI providers by updating AI_CONFIG.CURRENT_PROVIDER in utils/constants.js:

CURRENT_PROVIDER: AI_PROVIDERS.CHATGPT  // or OLLAMA, CLAUDE