✨ envite BPM AI Hackday: Integrating BPM with AI

Elevating Business Process Management (BPM) by integrating it with Artificial Intelligence (AI) and Machine Learning (ML).

In this project, we utilize Camunda 8 to orchestrate an AI-driven workflow. Our objective is to create connectors and establish BPMN patterns that harness the capabilities of machine learning and large language models, enabling the safe automation of such workflows.

🔭 Overview

This repository hosts the preconfigured backend service for the envite BPM AI Hackday. Camunda 8 operates within Docker containers, providing a flexible environment for orchestrating AI-driven workflows. The Spring Boot backend service harnesses the power of the Langchain4J Framework to integrate cutting-edge AI capabilities. It can either be launched directly from the IDE or deployed as a Docker container.

📙 Prerequisites

• The OpenAI API key that has been provided
• Docker: https://docs.docker.com/get-docker/
• Camunda Modeler: https://camunda.com/de/download/modeler/

🚀 Quick Start Guide

Enter your OpenAI API key in the application.yaml file located in the src/main/resources directory. The key should be assigned to the langchain4j.open-ai.chat-model.api-key property.

To build the backend application, execute mvn clean install in the root directory. Subsequently, run docker-compose up --build to containerize the backend service and initiate Camunda 8.

You can now access the various web applications. Use the username demo and the password demo to log in.

• Operate: http://localhost:8081
• Tasklist: http://localhost:8082
• Elasticsearch: http://localhost:9200

🔍 Evaluation

To evaluate the basic functionality, you can interact with the following test endpoints:

• GET http://localhost:8700/test
• GET http://localhost:8700/assistant

Additionally, you have the ability to initiate a new basic process instance in Operate and successfully accomplish the generated UserTask.

Please ensure that the backend service is either executed from the IDE or deployed as a Docker container, as both utilize the same port for communication.

💡 Useful Links

• Camunda AI Overview
• Camunda AI Process Enhancement
• Langchain4J Intro
• Introduction to LLMs and LangChain4J
• Langchain4J Repo
• Langchain4J Examples Repo ⭐
• Langchain4J Spring Examples Repo
• Tips for Promt Engineering
• Camunda Platform Repo