Confluent Unveils Streaming Agents to Power Real-Time Agentic AI with Flink

Confluent Inc., the data streaming leader behind Apache Kafka, has launched Streaming Agents, a new capability within Confluent Cloud for Apache Flink, aimed at scaling real-time, production-ready agentic AI applications.

The feature targets industries like financial services, healthcare, and advertising, where time-sensitive decision-making is critical.

Streaming Agents allow developers to build autonomous agents that adapt and act instantly on live data. This marks a shift from experimental generative AI use cases to real-world implementations with tangible ROI.

Built on Kafka and Flink, Streaming Agents support long-running workflows, secure integrations via Model Context Protocol (MCP), and access to APIs, databases, and vector stores. This lets enterprises deploy responsive agents for use cases like anti-money-laundering, RAG, and dark launches—bridging the gap between data, decisions, and action.

“Agentic AI has become an important element of competitiveness, but what separates enterprises getting real ROI from [foundation models] is the ability to contextualize AI with domain-specific data in real time,” said Andrew Sellers, Head of Technology Strategy at Confluent.

Sellers emphasised the gap between AI model capabilities and the availability of trustworthy, real-time data. “[Large language models] aren’t the inhibiting force anymore. What’s missing is contextual, trustworthy, real-time data that agents can use to make decisions,” he added.