Multi-Agent Customer Support System

This project demonstrates a multi-agent architecture using Azure OpenAI and Rust to enhance customer support operations. The system automates responses to common queries, escalates complex issues to humans, and analyzes customer sentiment to navigate.

Features

1. Automated Query Handling: AI agent handles common customer queries automatically.
2. Issue Escalation: Detects complex or sensitive issues and escalates them by deciding for action items or upper management escalation.
3. Sentiment Analysis: Analyzes customer sentiment to prioritize support tickets.

Components

• Customer Query Agent: Handles incoming customer queries and provides immediate responses using Azure OpenAI.
• Escalation Agent: Monitors interactions and escalates complex issues to humans or proposes action items.
• Sentiment Analysis Agent: Analyzes the sentiment of customer messages to identify unhappy customers.

Prerequisites

• Rust (latest version)
• An Azure OpenAI API key

Installation

1. Clone the repository

   git clone https://github.com/dminkovski/customer-support-assistant-rust.git
   cd customer-support-assistant-rust

2. Set up environment variables Create an .env file in the root directory and add your Azure OpenAI API key or copy the .env.sample.

   AZURE_OPEN_AI_ENDPOINT=https://XXXXXXXXX.openai.azure.com/
   AZURE_OPEN_AI_KEY=XXXXXXXXXXXXX
   AZURE_OPEN_AI_MODEL_DEPLOYMENT_NAME=gpt-4o
   AZURE_OPEN_AI_API_VERSION=2024-02-15-preview

3. Build the project

   cargo build

Usage

1. Run the project

   cargo run

SupportCase Struct

The SupportCase struct is used to log information and maintain a complete history of the interaction.

struct SupportCase {
    pub case_id: Uuid,
    pub support_context: String,
    pub customer_query: String,
    pub support_response: Option<String>,
    pub sentiment: Option<String>,
    pub should_escalate: bool,
    pub escalated: bool,
    pub needs_upper_management_attention: bool,
    pub created_at: DateTime<Local>,
    pub updated_at: DateTime<Local>,
    pub trace: Vec<Message>
}