Couchbase has announced the general availability of its AI Data Plane, a unified data infrastructure designed to help enterprises move AI agents from pilot projects to production by providing persistent memory, real-time context retrieval and governed data access across cloud, edge and lakehouse environments.
The company said the platform addresses one of the biggest barriers to enterprise AI adoption, the fragmented data infrastructure that often forces organisations to combine separate vector databases, caching systems and document stores to support AI agents. The AI Data Plane brings these capabilities together into a single operational layer, enabling enterprises to build more reliable and scalable agentic AI applications.
The platform combines Agent Memory, an Agent Catalog for discoverable agent tools, and an enterprise-supported Model Context Protocol (MCP) server, while supporting deployment across both Couchbase Capella and self-managed environments. It also integrates with lakehouse architectures through Enterprise Analytics 2.2, which introduces Apache Iceberg federation and a forthcoming Trino adapter.
According to Couchbase, enterprises increasingly require AI agents capable of retaining conversational context, accessing structured operational data and maintaining state across sessions and capabilities that extend beyond traditional vector search.
The company said its Agent Memory capability provides a unified persistence layer for conversational context, operational data and memory retrieval, while remaining compatible with orchestration frameworks including LangGraph, CrewAI and LlamaIndex.
The announcement also includes enhancements to Enterprise Analytics 2.2, allowing organisations to query operational data alongside Apache Iceberg lakehouse tables without duplicating data through complex ETL processes. Additional analytics improvements include support for Google Cloud Storage, JWT authentication, Oracle and SQL Server change data capture, asynchronous queries and expanded SDK capabilities.
For AI workloads deployed beyond centralised cloud environments, Couchbase has extended the AI Data Plane to edge and mobile deployments through updates to Couchbase Lite, Edge Server, React Native support and Sync Gateway. The company said these capabilities allow AI agents running on devices and edge environments to access replicated data and perform local vector searches even in low-connectivity scenarios.
Barry Morris, Chief Product and Strategy Officer at Couchbase, said the company developed the AI Data Plane in response to customer demand for a unified data layer that eliminates the complexity of managing separate vector, caching and document stores for production AI agents. He added that the platform is designed to provide a framework-agnostic persistence layer capable of operating consistently across cloud, self-managed and edge environments.