Cloud-GPT: Enterprise Java Meets Generative AI

A demo application showcasing the integration of Jakarta EE 10, MicroProfile, Payara Server/Cloud, and OpenAI's GPT API to build an intelligent travel recommendation system.

Technologies

• Jakarta EE 10
• MicroProfile Config & Cache
• Payara Server 6
• Payara Cloud deployment support
• OpenAI GPT API integration

Prerequisites

• Java 21+
• Payara Server 6.2024.10 or later
• Docker

Configuration

1. Create microprofile-config.properties:

openai.key=${OPENAI_API_KEY}

Deployment Options

Local Development (Payara Server)

mvn clean package
docker compose up -d

Payara Cloud

1. Push to your GitHub repository
2. Connect repository to Payara Cloud
3. Deploy with a single click

Features

• GPT-powered travel recommendations
• Distributed caching via Payara Data Grid
• MicroProfile Config for external configuration
• Cloud-native deployment ready

Architecture

• Jakarta EE REST endpoints
• GenAI integration via OpenAI client
• JCache for response caching
• Containerized deployment

Access the application at: http://localhost:8080/jee-chatgpt You can also access the app deployed to Payara Cloud at Jakarta-101

Note: Initial GPT responses may take 30+ seconds. Subsequent identical queries are served from cache.