Confluent has introduced the Streamhouse data architecture to give AI agents a continuously updated and governed view of business operations.

Unveiled at Current San Francisco 2026, Streamhouse continuously captures, processes, governs, and serves real-time business context while working alongside existing data warehouses and lakehouses.

To support it, Confluent has added query capabilities to its data streaming platform, which now covers stream, connect, process, govern, and query.

“The next era of enterprise AI will be won by the agents that know what’s happening in the business the moment they act,” said Shaun Clowes (top), General Manager of Confluent and Data at IBM.

Lightning Tables, now generally available on AWS, let applications query live streaming data with low latency and without separate databases, caches or serving systems.

A Real-Time Context Engine delivers current, governed context to AI agents through the Model Context Protocol. Confluent Tableflow then makes the same streaming data available to existing lakehouses and warehouses for analytics. That links real-time operations with the wider enterprise data estate.

Across Asia-Pacific, businesses are scaling digital services quickly and customers expect experiences that respond in the moment.

“A Streamhouse gives teams a continuously updated, governed foundation for that context, so applications and agents can work from the same current picture,” said Greg Taylor, SVP and General Manager of APAC at Confluent.