Introducing the InterSystems IRIS Document Store for Haystack

Artificial Intelligence applications are increasingly built around Retrieval-Augmented Generation (RAG), semantic search, and AI agents. As these applications move into production, choosing the right persistence layer becomes just as important as selecting the LLM.

Today, I'm excited to announce the InterSystems IRIS Document Store for Haystack, a new open-source integration that enables developers to use InterSystems IRIS as a native Document Store within the Haystack AI framework.

Why Haystack?

Haystack has become one of the leading open-source frameworks for building production-ready AI applications. Its modular pipeline architecture makes it easy to create solutions for:

• Retrieval-Augmented Generation (RAG)
• Enterprise Search
• Question Answering
• AI Agents
• Semantic Search
• Knowledge Assistants

Introducing the Integration

The InterSystems IRIS Document Store implements Haystack's Document Store interface, allowing it to integrate naturally into existing Haystack pipelines.

Whether you're building a small proof of concept or a production RAG system, switching to IRIS as your persistence layer requires minimal changes to your application.

Installation

The package is available on PyPI.

pip install intersystems-iris-haystack

Resources

Official Haystack Integration

https://haystack.deepset.ai/integrations/intersystems-iris-document-store

GitHub Repository

https://github.com/s-c-ai/iris-haystack

PyPI Package

https://pypi.org/project/intersystems-iris-haystack/

Example Architecture

                 Haystack Pipeline

Converter → Splitter → Embedder

                │
                ▼

      InterSystems IRIS Document Store

      • Documents
      • Metadata
      • Vector Search
      • SQL
      • Objects

                │
                ▼

      Retriever → LLM → Answer