{"slug": "couchbase-extends-database-reach-to-ai-agents", "title": "Couchbase Extends Database Reach to AI Agents", "summary": "Couchbase added a data plane to its database platform to give AI agents persistent memory access for context retrieval, reducing token usage and costs. The company also introduced an MCP server, Agent Catalog, and updates to its analytics, edge, and mobile offerings to support AI workloads.", "body_md": "Couchbase today added a data plane to its database platform to provide artificial intelligence (AI) agents with access to persistent memory that provides the context needed to more cost-effectively automate tasks.\n\nAdditionally, Couchbase has added a Model Context Protocol (MCP) server and an Agent Catalog through which AI agents can be discovered.\n\nAt its core, the Couchbase database is based on a JavaScript object notation (JSON) format that is designed to run in memory to process data with sub-millisecond latency at scale. That ability to instantly access data is now being extended to AI agents to enable them to retrieve context and, when needed, write data across thousands of concurrent sessions in a way that preserves state.\n\nCouchbase today also updated its Enterprise Analytics platform to add support for the Iceberg table format advanced by the Apache Foundation, along with a Trino adapter, due out in the third quarter, through which SQL can be used to access Couchbase from other platforms that are based on the Trino distributed query engine.\n\nOther extensions to the Enterprise Analytics platform include support for Google Cloud Storage, JWT authentication, Oracle and SQL Server change data capture, asynchronous long-running queries, an index advisor, index-only query plans, SQL++ UPDATE support, and updates to the Java, .NET, Python, and JavaScript software development kits (SDKs).\n\nCouchbase has extended its managed Capella to make multiple types of AI models accessible to the natural language query assistant through which IT teams can also apply policies to how data is accessed.\n\nAt the same time, Couchbase has updated the Couchbase Lite (CBL) database for edge computing applications to add Bluetooth support alongside existing support for WiFi networks, along with an Android application programming interface (API) with native Kotlin @Serializable support to eliminate boilerplate data mapping and new C++ API bindings to simplify building applications.\n\nAn Edge Server 1.1 update, meanwhile, adds support for fine-grained local permissions, CORS support for browser-based edge apps, simplified credential rotation for distributed device fleets, and expanded platform support for Windows and ARM architectures.\n\nFinally, Couchbase is adding a React Native 1.1 update to support Turbo Module integration to give IT teams direct access to Couchbase Lite performance without bridging overhead, while a Sync Gateway 4.1 has been updated to enable non-disruptive rolling upgrades and concurrent distributed resynchronization for high-volume workloads,\n\nThe overall goal is to make data more accessible to AI applications and agents in a way that ultimately serves to reduce the total number of tokens that need to be consumed, says Tim Finley, senior vice president for AI at Couchbase.\n\nThe Couchbase AI Data Plane, in effect, creates a unified persistence layer that reduces the amount of data that an AI agent would otherwise need to include within a context window that consumes memory that could be better allocated to reasoning tasks, said Finley. “That results in lower token usage,” he adds.\n\nThe result should be higher-quality outputs that are provided at a significantly lower cost using a multimodal platform designed to horizontally scale, notes Finley.\n\nOrganizations can also more easily switch between AI models and data sources using an agnostic framework that can be validated using frameworks such as LangGraph, CrewAI, and LlamaIndex, he says.\n\nThere is, of course, no shortage of ways to provide AI agents with the context they need to automate a task. The challenge and the opportunity now are determining which approach ultimately achieves that goal in the most efficient way possible.", "url": "https://wpnews.pro/news/couchbase-extends-database-reach-to-ai-agents", "canonical_source": "https://techstrong.ai/features/couchbase-extends-database-reach-to-ai-agents/", "published_at": "2026-06-30 13:00:24+00:00", "updated_at": "2026-06-30 13:30:34.207799+00:00", "lang": "en", "topics": ["ai-agents", "ai-infrastructure", "artificial-intelligence"], "entities": ["Couchbase", "Tim Finley", "Apache Foundation", "Google Cloud", "Oracle", "SQL Server", "LangGraph", "CrewAI"], "alternates": {"html": "https://wpnews.pro/news/couchbase-extends-database-reach-to-ai-agents", "markdown": "https://wpnews.pro/news/couchbase-extends-database-reach-to-ai-agents.md", "text": "https://wpnews.pro/news/couchbase-extends-database-reach-to-ai-agents.txt", "jsonld": "https://wpnews.pro/news/couchbase-extends-database-reach-to-ai-agents.jsonld"}}