Why We Keep Process Data Outside the Engine - and Why It Changes Everything for Agentic BPM

The article introduces "in-concert," an open-source BPMN engine designed specifically for building agentic systems, which are applications that combine AI capabilities with traditional business processes. Here's a summary of the key points:

1. Purpose: In-concert aims to provide a flexible and scalable solution for orchestrating complex workflows in agentic systems by leveraging BPMN (Business Process Model and Notation).

2. Architecture:

   • The engine is designed to work with external data sources, allowing users to manage their own databases or use any other storage system.
   • It separates the orchestration logic from business intelligence, enabling developers to integrate AI services without tightly coupling them to the engine's internal state.

3. Key Features:

   • Service Tasks: Service tasks in BPMN diagrams can be implemented as calls to external AI services (e.g., LLMs).
   • Gateway Decisions: Gateway decisions are evaluated by custom logic or AI models, rather than being hardcoded within the engine.
   • Human Tasks: Human tasks are managed through a worklist API, allowing for flexible user interaction and task assignment.

4. Implementation Details:

   • The engine can be run as a standalone service (via REST

Read the full article at DEV Community

---

Want to create content about this topic? Use Nemati AI tools to generate articles, social posts, and more.