RavenDB Launches Innovative Native DBMS AI Agents

 

 

RavenDB Launches Innovative Native DBMS AI Agents

By Betsy Burton

September 8, 2025, RavenDB announced the launch of its new AI Agent Creator, a feature fully integrated into its multi-model NoSQL document database.

RavenDB is a native AI database that supports ACID transactions without SQL. It is designed to specifically support AI workload.

By bringing the AI agent functionality directly into the database, RavenDB aims to remove the common hurdles associated with integrating AI models with complex data systems, a problem that often causes enterprise AI projects to fail.

What Are RavenDB Innovative Native DBMS AI Agents?

RavenDB’s AI Agent Creator allows developers to build and manage context-aware AI agents natively inside the database. These DBMS stored agents are reminiscent of old DBMS stored logic. However, these agents can dynamically handle natural language requests, query live operational data, and trigger defined actions, all within a secure, controlled environment.

A key innovation is the “zero-trust, default-deny” security model for the LLMs, which ensures that no data or operations are accessible to the agent unless explicitly approved by the developer. This provides a strong security framework and gives developers precise control over how agents interact with the data.

The feature, part of the new RavenDB 7.1 release, also supports all major LLMs and includes smart caching to optimize performance and reduce AI-related costs.

Why are Native DBMS AI Agents Important?

This announcement illustrates a trend I am seeing with other AI-focused databases of embedding AI and machine learning capabilities directly into the database. We have discussed this with other products such as Databricks Datalake.

The key differentiator is that RavenDB’s AI agents are not an external connector – but rather stored and processed with the data. This could significantly enhance RavenDB’s value proposition for enterprises looking to build intelligent applications without the typical integration overhead.

While the in-database approach offers substantial benefits in speed and security, it also introduces a new set of challenges, such as monitoring and performance tuning for the embedded AI agents, which will require careful consideration from both developers and database administrators.

Why are Native DBMS AI Agents So Powerful?

RavenDB’s DBMS AI Agent Creator is a framework for building and running intelligent agents that operate directly on the data. They are not just executing pre-defined scripts; they are orchestrating complex, multi-step workflows in response to natural language and using an LLM as their “brain.”

• Agentic and dynamic. The agent’s core function is to use its reasoning capabilities (powered by an LLM) to decide which tools to use and in what order to achieve its goal. It’s not a pre-written script; it’s a decision-making entity.
• Input/Output: Their primary input is often a natural language prompt from a user (“Find all customers who bought a specific product and have a high risk score”). The agent’s output can be a direct answer, a summary, or the execution of a series of actions.
• Intelligence: They are intelligent. They use an LLM to understand the user’s intent, reason about the best course of action, and dynamically generate the steps needed to fulfill the request.
• Debugging and Maintenance: RavenDB’s system is designed to provide better guardrails and control. You define the “tools” and the scope, and the LLM reasons within those boundaries, which is a more manageable approach than trying to write every possible script.

Impact on the Market

RavenDB is in an enviable position as they do not have a legacy of SQL-based applications and tools. As such, it can support these capabilities natively which can lower the barrier to entry for firms that have previously found AI implementation too complex or risky.

This is another case of AI-native DBMSs shifting the from being a repository to being that of an intelligent, operational platform capable of executing complex AI workflows directly on the data.

Bottom Line

The RavenDB Native DBMS AI Agents is a significant innovation that illustrates a broader trend toward embedding AI capabilities and agents natively in their DBMS. This move reinforces the growing importance of a “data-first” approach to AI development, where the database becomes the central hub for both data and intelligence.

Clients should evaluate this native DBMS AI agent technology as a way to accelerate their AI initiatives, reduce development time, and mitigate security risks associated with data movement.

 

Our Third Transform Tour for 2025 featuring our Predictions for 2026.

Join  Betsy Burton, Jim Lundy and Adam Pease to hear where markets are going over the next three years.

Register Today