Customer Assistant

This application is a customer assistant that can answer questions and also take actions about and towards a customer's account.

Features

This application utilizes the following features:

• Semantic Kernel for Java
  • Vector data stores
  • Chat completions with OpenAI
  • Embedding with OpenAI
• OpenAI
• React UI
• Quarkus

Getting Started

Deploy To Azure Container Apps

To deploy the application you will need a deployment of OpenAI. This will need to have 2 models deployed to it:

• A chat completion model, by default this is named gpt-4o but this can be configured.
• An embedding model, by default this is named text-embedding-3-large but this can be configured.

You will be asked for the resource group and the name of the OpenAI instance during deployment.

## If needed
# azd auth login

azd up

You can now browse to the web service of the deployed asset that will be displayed in the output of the azd up. Will look like: https://ca-web-XXXX.azurecontainerapps.io/

Run Locally

Before running the application:

1. In the docker folder, make a copy of demo.properties_example to demo.properties (in the same folder), and fill in the required keys.

To run the application locally as a container, have Docker Desktop or Podman installed, then run the following command:

cd docker
docker compose up --build --force-recreate

When started the app will be available at http://localhost:3000/.

Resources

• Semantic Kernel for Java
• Semantic Kernel for Java Documentation

Guidance

Costs

This application uses Azure OpenAI, which may incur costs. Please refer to the Azure OpenAI website for more information.

License

This project is licensed under the MIT License - see the LICENSE file for details.