dc-customer-support-agent

📖 Overview

Welcome to the dc-customer-support-agent project! This practical example accompanies an article about AI agents powered by LangGraph in InterSystems IRIS.

The project demonstrates an AI-based customer support agent capable of analyzing incoming email requests, determining their priority, and categorizing them appropriately.

🛠️ How It Works

The Customer Support AI Agent is designed to automate the initial handling of customer support emails. Its workflow is as follows:

Read Incoming Support Emails:
- The agent accesses emails requiring assistance.
Classify Priority:
- Emails are analyzed to determine their priority level: High, Medium, or Low.
Identify Topic:
- The agent detects the topic or category of the request, such as password reset, VPN issues, printer problems, etc.
Decision Making:
- The agent decides whether to auto-respond to the email or escalate it to human support for further attention.
Auto-Responding:
- If auto-responding is the chosen course of action, the agent retrieves past examples (RAG) and crafts a personalized reply.

📋 Prerequisites

Before you begin, ensure you have the following:

Docker and Docker Compose installed on your system.
An .env file set up with necessary environment variables. You can use the env_sample file available at the repository’s root directory as a template.

🛠️ Installation

Follow these steps to set up the project:

Clone the repository:

git clone https://github.com/henryhamon/dc-customer-support-agent

Navigate to the project directory:
```
cd dc-customer-support-agent
```
Configure environment variables:
- Copy the env_sample file to a new file named .env and fill in the required details.

Build the Docker container:

docker-compose build --no-cache --progress=plain

💡 How to Use

To start and manage the Customer Support AI Agent, use Docker Compose:

Start the Application:
- Run the following command to start the application in detached mode:
```
docker-compose up -d
```
Stop and Remove Containers:
- To stop the application and remove containers along with their images, use:
```
docker-compose down --rmi all
```