AI-Powered Support System

A hybrid AI-powered customer support system that combines OpenAI's GPT-4 for complex queries with local LLM (Ollama) for simple queries. The system includes sentiment analysis, real-time chat, and a knowledge base for efficient customer support.

Features

• Hybrid AI System: Combines OpenAI's GPT-4 for complex queries with local LLM (Ollama) for simple queries
• Intelligent Routing: Automatically routes queries to the appropriate AI model based on complexity
• Sentiment Analysis: Analyzes customer sentiment and escalates to human agents when needed
• Real-time Chat: WebSocket-based real-time communication with typing indicators and read receipts
• Knowledge Base: FAQ system with automatic suggestions based on conversation context
• User Authentication: JWT-based authentication with role-based access control
• Admin Dashboard: Analytics and management interface for administrators
• Conversation History: Search and review past conversations
• Scalable Architecture: Built with Go, PostgreSQL, and Redis for high performance

Technology Stack

Backend

• Go: Programming language
• Gin: Web framework
• PostgreSQL: Primary database
• Redis: Caching and session management
• JWT: Authentication
• GORM: ORM for database operations
• Logrus: Structured logging
• Viper: Configuration management

AI Integration

• OpenAI API: For complex queries
• Ollama: For local LLM integration
• Sentiment Analysis: For analyzing customer sentiment

Frontend (To be implemented)

• React/Vue.js: Frontend framework
• WebSocket: Real-time communication
• Material-UI/Ant Design: UI components

Deployment

• Docker: Containerization
• Docker Compose: Local development
• Kubernetes: Production deployment
• CI/CD: Automated testing and deployment

Project Structure

support/
├── backend/                 # Backend Go application
│   ├── cmd/                # Application entry point
│   │   └── main.go        # Main application file
│   ├── internal/          # Internal application packages
│   │   ├── ai/            # AI service
│   │   ├── auth/          # Authentication service
│   │   ├── conversation/  # Conversation service
│   │   ├── database/      # Database connection and migrations
│   │   ├── handlers/      # HTTP handlers
│   │   ├── knowledge/     # Knowledge base service
│   │   ├── models/        # Data models
│   │   └── routes/        # Route definitions
│   ├── pkg/               # Public library packages
│   │   ├── config/        # Configuration management
│   │   └── logger/        # Logging utilities
│   ├── go.mod             # Go module file
│   ├── go.sum             # Go module checksums
│   └── Dockerfile         # Docker configuration for backend
├── frontend/              # Frontend React/Vue.js application (to be implemented)
├── docs/                  # Documentation
├── deployment/            # Kubernetes deployment configurations (to be implemented)
├── docker-compose.yml     # Docker Compose configuration
├── .gitignore             # Git ignore file
└── README.md              # This file

Getting Started

Prerequisites

• Go 1.21 or higher
• Docker and Docker Compose
• PostgreSQL (if not using Docker)
• Redis (if not using Docker)
• OpenAI API key (for GPT-4 integration)

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/support.git
   cd support
   

2. Set up environment variables:

   cp backend/.env.example backend/.env
   

   Edit the .env file with your configuration, including your OpenAI API key.

3. Start the services using Docker Compose:

   docker-compose up -d
   

4. Wait for all services to start up. You can check the logs with:

   docker-compose logs -f
   

Development

Backend Development

1. Navigate to the backend directory:

   cd backend
   

2. Install dependencies:

   go mod download
   

3. Run the application:

   go run cmd/main.go
   

The backend API will be available at http://localhost:8080.

Frontend Development (To be implemented)

1. Navigate to the frontend directory:

   cd frontend
   

2. Install dependencies:

   npm install
   

3. Run the development server:

   npm run dev
   

The frontend will be available at http://localhost:3000.

API Documentation

Authentication Endpoints

• POST /api/v1/public/login - User login
• POST /api/v1/public/register - User registration

User Endpoints

• GET /api/v1/user/profile - Get user profile
• PUT /api/v1/user/profile - Update user profile
• PUT /api/v1/user/change-password - Change password

Admin Endpoints

• GET /api/v1/admin/users - List all users
• GET /api/v1/admin/users/:id - Get user by ID
• PUT /api/v1/admin/users/:id - Update user
• DELETE /api/v1/admin/users/:id - Delete user

Conversation Endpoints

• GET /api/v1/conversations - List conversations
• POST /api/v1/conversations - Create new conversation
• GET /api/v1/conversations/:id - Get conversation by ID
• PUT /api/v1/conversations/:id - Update conversation
• DELETE /api/v1/conversations/:id - Delete conversation
• GET /api/v1/conversations/:id/stats - Get conversation statistics

Message Endpoints

• GET /api/v1/conversations/:id/messages - Get messages in a conversation
• POST /api/v1/conversations/:id/messages - Send a message
• PUT /api/v1/conversations/:id/messages/:messageId - Update a message
• DELETE /api/v1/conversations/:id/messages/:messageId - Delete a message
• POST /api/v1/conversations/:id/ai - Send message with AI response

Knowledge Base Endpoints

• GET /api/v1/knowledge - Get knowledge base entries
• GET /api/v1/knowledge/search - Search knowledge base
• GET /api/v1/knowledge/categories - Get knowledge categories
• GET /api/v1/knowledge/tags - Get knowledge tags
• GET /api/v1/knowledge/popular - Get popular knowledge entries
• GET /api/v1/knowledge/recent - Get recent knowledge entries
• GET /api/v1/knowledge/best-match - Find best match for a query
• GET /api/v1/knowledge/stats - Get knowledge base statistics
• GET /api/v1/knowledge/:id - Get knowledge base entry by ID
• POST /api/v1/knowledge/:id/rate - Rate knowledge base entry

Admin Knowledge Base Endpoints

• POST /api/v1/admin/knowledge - Create knowledge base entry
• PUT /api/v1/admin/knowledge/:id - Update knowledge base entry
• DELETE /api/v1/admin/knowledge/:id - Delete knowledge base entry

AI Endpoints

• POST /api/v1/ai/query - Query AI service
• POST /api/v1/ai/analyze-complexity - Analyze complexity of a prompt
• GET /api/v1/ai/models - Get available AI models
• POST /api/v1/ai/openai - Query OpenAI directly
• POST /api/v1/ai/ollama - Query Ollama directly

Testing

Backend Testing

1. Navigate to the backend directory:

   cd backend
   

2. Run tests:

   go test ./...
   

Frontend Testing (To be implemented)

1. Navigate to the frontend directory:

   cd frontend
   

2. Run tests:

   npm test
   

Deployment

Docker Deployment

1. Build the Docker image:

   docker build -t support-backend ./backend
   

2. Run the container:

   docker run -p 8080:8080 support-backend
   

Docker Compose Deployment

1. Start all services:

   docker-compose up -d
   

2. Stop all services:

   docker-compose down
   

Kubernetes Deployment (To be implemented)

1. Apply the Kubernetes configurations:

   kubectl apply -f deployment/