Database platform provider Couchbase has unveiled its AI Data Plane, a new unified data infrastructure designed to help enterprises deploy and manage artificial intelligence (AI) agents at scale by providing persistent memory, real-time context retrieval, and consistent data access from cloud to edge environments.

In a statement, Couchbase said the platform aims to address data management challenges that often hinder the transition of AI projects from pilot programs to full-scale production deployments. The AI Data Plane combines operational data, context, and memory into a single governed layer, enabling AI agents to make more consistent decisions and deliver improved customer experiences.

The company said the platform integrates several capabilities, including Agent Memory, an Agent Catalog for discovering agent tools, a self-managed Model Context Protocol (MCP) server, and an LLM cache designed to reduce redundant inference requests. These components operate across both Couchbase’s cloud-based Capella platform and self-managed deployments.

According to Couchbase, the platform is intended to simplify enterprise AI infrastructure by replacing multiple standalone systems used for caching, vector search, and document storage with a single operational framework.

Research firm IDC said real-time and contextual data will be essential for most agentic AI applications in the coming years, making data architecture a critical factor in successful deployments.              

Couchbase also announced Enterprise Analytics 2.2, which introduces Apache Iceberg lakehouse federation and a new Trino adapter.