Confluent adds Intelligence, Cloud features

Confluent said on May 19 that it added new capabilities in Confluent Intelligence and Confluent Cloud aimed at helping companies build and secure real-time AI applications. The company said the release includes tools for agent-powered workflows, automated data protection and private cloud connectivity. Confluent described the update as part of its push to make streaming data usable as live context for AI systems rather than as a batch input delivered after the fact. The new package centers on features Confluent has been rolling out across its managed cloud platform. Confluent Cloud’s Q2 2026 launch includes a dbt adapter for Flink, a managed Model Context Protocol server, Confluent Agent Skills, and general availability for both Streaming Agents and the Real-Time Context Engine, according to the company’s May blog post. Confluent said those tools are meant to let developers use SQL workflows, governed event streams and built-in AI functions inside one service. (businesswire.com) ### Which new pieces are now available? Streaming Agents and the Real-Time Context Engine are now generally available, according to Confluent’s May 2026 product updates. Confluent said Streaming Agents let users build, deploy and orchestrate event-driven agents natively on Apache Flink and Kafka, while the Real-Time Context Engine continuously enriches operational events with external data and built-in AI or machine-learning functions. (confluent.io) Confluent also added what it called expanded agent operations, more model options and new ML functions for real-time AI in a separate Q2 Confluent Intelligence update published this week. The company said those additions are designed to give AI applications “fresh, governed context” without requiring customers to stitch together separate infrastructure. ### How does this connect to LLM and agent workflows? (confluent.io) The managed MCP server and Agent Skills are the developer-facing pieces most directly tied to large language model workflows. Confluent said the managed MCP server brings Confluent capabilities into AI coding agents, while Agent Skills package Confluent best practices for those tools. Confluent’s documentation says Streaming Agents can invoke tools to interact with external systems, retrieve information and perform actions as part of an AI workflow. (confluent.io) The company said those agents operate inside data streams so they can monitor business events and respond in real time, rather than waiting for a separate application layer to poll for updates. ### What changed in Confluent Cloud itself? (confluent.io) Confluent said the broader cloud release also included private networking and governance features beyond the AI layer. In its May 19 announcement, the company said the update adds automated data protection and private cloud connectivity as part of the same release cycle. The release notes say customers can now create, manage and monitor Streaming Agents directly in the Confluent Cloud Console. (docs.confluent.io) Confluent also said the AI_DETECT_PII built-in function is available in early access, adding another control aimed at handling sensitive data in AI pipelines. ### Where do customers get the new tools? Confluent’s documentation says the features are available through Confluent Cloud, with agent management exposed in the cloud console. (businesswire.com) The company’s release notes and product pages direct users to the console and documentation for setup, monitoring and deployment. Confluent’s next visible milestone is customer adoption of the Q2 2026 release set through Confluent Cloud, where Streaming Agents, Real-Time Context Engine and related tools are now listed as available or generally available in current documentation. (docs.confluent.io) (confluent.io)