Confluent Adds Query Layer and Lightning Tables

Confluent says the Streamhouse is designed to keep business context current across agents, applications and analytics, rather than forcing firms to stitch together batch warehouses, data lakes and separate serving systems. The platform now spans Stream, Connect, Process, Govern and Query, with Query providing the access layer for analytics, AI agents and applications.

The new Lightning Tables feature is central to that push. Available on Confluent Cloud on AWS, it materializes Kafka data into read-only tables that stay current as changes arrive, with Confluent citing internal testing that shows about 100 milliseconds from data creation to serving, 10 to 100 milliseconds of query latency and throughput above 1,000 queries per second.

That matters for investors because Confluent is trying to turn its Kafka franchise into a broader enterprise platform with higher switching costs and more monetization points. If customers adopt its query and serving tools alongside streaming and governance, that could deepen usage inside large accounts and make Confluent more central to AI infrastructure spending.

The company is also leaning on customer validation. Accenture said moving to a Streamhouse decouples data producers from consumers and has cut time to insight from months to days, while Extend said the architecture helps with real-time pricing and instant fraud detection. Both use cases highlight where enterprises are likely to spend next: live decisioning, compliance and personalization rather than static reporting.

The timing fits a wider shift in AI adoption. A recent study suggested AI tools can lengthen workdays rather than shorten them, underscoring how much enterprise value still depends on making AI more accurate and more operationally useful. Confluent’s pitch is that live, governed data is what turns AI from a productivity experiment into a production system.

For investors, the key question is whether the Streamhouse narrative translates into durable demand and larger deals. The next catalyst will be whether Confluent can convert the architecture story into measurable customer expansion and whether enterprises treat real-time data infrastructure as a must-have layer for AI agents rather than another optional add-on.