RavenDB introduces Quill, a context layer that sits on top of existing PostgreSQL, SQL Server, and MySQL databases. This lets organizations deploy AI agents in production without migrating their system of record or building a complete AI stack.
RavenDB, a NoSQL document database, aims to address a problem many IT departments now face with Quill. Business-critical data resides in SQL systems built years before the rise of embeddings and agents. Integrating AI applications into these systems is difficult.
Modernizing or replacing those systems is expensive and risky, RavenDB says. Quill takes a different approach: it adds a context layer on top of the existing database, which remains in place. The package includes search, retrieval, and the agents themselves. According to the company, production-ready agents can go live in a matter of weeks, compared with the 18 to 24 months a typical in-house build would take.
Quill sits between the model and the source system, based on the principle that the model does not deserve unlimited access. Organizations determine what an agent is and isn’t allowed to see, independent of the underlying database permissions. The example RavenDB provides: a healthcare agent that answers questions about an upcoming appointment, while prescription data is never part of the dataset it is allowed to search.
Model-agnostic, cloud or on-premises
Founder and CEO Oren Eini emphasizes the work behind the demo. “Anyone can stand up an AI demo in an afternoon, but getting that demo into production with data pipelines, semantic search, security, governance, all the plumbing a small proof of concept doesn’t need until it has to run at scale, is the hard part,” says Eini. According to him, Quill provides teams with that infrastructure out of the box.
Quill is model-agnostic. Teams can use any AI model, switch providers, or run everything on their own hardware. Agents support web chat, WhatsApp, Telegram, Slack, and Discord out of the box.
The product is now available for PostgreSQL, SQL Server, and MySQL, with support for more databases on the horizon. You can deploy it in the cloud or on-premises to meet data residency and regulatory requirements.