Customer Service AI Trainer

This project is an AI-based chat platform for training Customer Service agents. The entire app is AI-powered and uses the OpenAI chat completion API to simulate a potential customer. The platform allows agents to respond to customer queries and provides them with immediate, real-time feedback, which they can use to improve their responses. The platform is built using FastAPI and ReactJS and is containerized using Docker.

See the demo below:

The project code can be accessed in the backend-app and frontend-app directories. See the directory structure below:

├── docs
├── backend-app
├── frontend-app
└── docker-compose.yml

Technologies

System Diagram

Setting up the environment

Prerequisites

Ensure you have Docker installed on your machine. If not, you can download it here.

Clone the app

$ git clone --recurse-submodules -b demo https://github.com/ensynAI/customer-service-ai-trainer.git

Environment variables

Create a .env file containing your OpenAI API Key in the root directory (devops).

.env file structure:

OPENAI_API_KEY="your-api-key"

Running the app

To launch the app run the following command. The command builds the docker images and launches the app (this may take some time).

$ docker-compose up

To rebuild everything from scratch use:

$ docker-compose up --build --force-recreate --no-deps

Running the app

To use the app go to localhost:3000

• Fronend app: localhost:3000

• Backend server: localhost:8000