Contact Center Agentic Flow System

An intelligent, multi-agent contact center system built with LangGraph and Vertex AI that provides automated customer support through specialized AI agents.

🚀 Quick Start

# Install dependencies
pip install -r requirements.txt

# Set up environment variables
cp .env.example .env
# Edit .env with your configuration

# Initialize database
python -m src.database.init_db

# Run the application
python -m src.main

🧪 Testing the Application

Method 1: Interactive Test Client (Recommended)

The easiest way to test customer interactions:

python test_customer_interaction.py

This provides:

• ✅ Automatic authentication
• ✅ Pre-built test scenarios
• ✅ Interactive conversation flow
• ✅ Real-time agent responses

Method 2: Manual API Testing

1. Start the Application

python -m src.main
# Application will run on http://localhost:8000

2. Get Authentication Token

curl -X POST "http://localhost:8000/token" \
  -H "Content-Type: application/x-www-form-urlencoded" \
  -d "username=johndoe&password=secret"

3. Start a Conversation

curl -X POST "http://localhost:8000/api/v1/conversations" \
  -H "Authorization: Bearer YOUR_TOKEN_HERE" \
  -H "Content-Type: application/json" \
  -d '{
    "customer_id": "CUST_001",
    "channel": "web",
    "initial_message": "I need help with my account",
    "priority": "medium"
  }'

4. Send Follow-up Messages

curl -X POST "http://localhost:8000/api/v1/conversations/CONVERSATION_ID/messages" \
  -H "Authorization: Bearer YOUR_TOKEN_HERE" \
  -H "Content-Type: application/json" \
  -d '{
    "content": "I cant access my billing information"
  }'

Test Credentials

• Username: johndoe
• Password: secret

Test Scenarios

Try these conversation starters to test different agent types:

1. General Support: "What are your business hours?"
2. Technical Issues: "My internet connection is very slow"
3. Billing Questions: "I was charged twice for the same service"
4. Sales Inquiries: "I'm interested in upgrading to premium"

🏗️ Architecture

• LangGraph Orchestration: Multi-agent workflow management
• Vertex AI Integration: Gemini Pro, Claude 3, and custom models
• FastAPI Backend: RESTful API with WebSocket support
• PostgreSQL: State management and analytics
• Redis: Caching and session management

📋 Features

• ⚡ Sub-30 second response times
• 🤖 6 specialized AI agents (Tier 1-3, Sales, Billing, Supervisor)
• 🔄 Intelligent routing and escalation
• 📊 Real-time monitoring and analytics
• 🛡️ Enterprise security and compliance
• 🌐 Multi-channel support (Web, Voice, Email, Mobile, Social)

🧪 Available Test Endpoints

• GET /health - Application health check
• POST /token - Get authentication token
• POST /api/v1/conversations - Start new conversation
• POST /api/v1/conversations/{id}/messages - Send message
• GET /api/v1/conversations/{id}/state - Get conversation state

📊 Monitoring

Check application status:

curl http://localhost:8000/health

Check system metrics (requires authentication):

curl -H "Authorization: Bearer YOUR_TOKEN" http://localhost:8000/metrics

🐛 Troubleshooting

Common Issues

1. Port already in use: The application runs on port 8000 by default. You mentioned it's listening on port 6379 - this might be a Redis instance.

2. Authentication errors: Make sure to use the correct credentials (johndoe/secret)

3. Database connection: Run python test_startup.py to verify all components are working

Debug Mode

export LOG_LEVEL=DEBUG
python -m src.main

📚 Documentation

For detailed testing instructions, see TESTING_GUIDE.md

🚀 Production Deployment

See deployment guide for production setup instructions.

---

Ready to test! 🎉 Run python test_customer_interaction.py to start your first conversation with the AI agents.