Dremio, the Intelligent Data Lakehouse Platform, is marking the next generation of its latest platform release, offering breakthrough intelligence capabilities that help organizations accelerate AI and analytics projects, and reduce costs.
Announced at Iceberg Summit 2025, the new Autonomous Reflections, Iceberg Clustering, AI-enabled Semantic Search, and enhancements to the Enterprise Catalog eliminate manual performance tuning and simplify data discovery.
By reducing technical complexity and resource waste through intelligent automation, Dremio liberates teams to innovate, all while strengthening data governance, according to the company.
With less time spent tuning performance, optimizing tables, and creating pre-calculated tables and extracts, data engineers can focus on delivering high-quality datasets that power agentic applications.
Dremio’s technology leverages intelligence to deliver:
- Autonomous Reflections—an industry-first innovation developed by Dremio—automatically optimizes performance while ensuring queries run on live, up-to-date data with no SQL changes.
- Iceberg Clustering—the industry’s first data clustering offering for Iceberg—brings automated data layout optimization to Apache Iceberg lakehouses.
Dremio is currently used by thousands of companies to develop a unified semantic layer, and with its next-generation release it is leveraging AI to make discovery across tables and sources easier with:
- AI-enabled Semantic Search—reduces data discovery time from days to seconds, even within complex data environments. Faster data discovery significantly accelerates AI and analytics initiatives by making it simple for users and AI agents to discover existing data assets across an enterprise's entire data estate using plain business language. It also removes the need for SQL experience and/or technical expertise.
The AI era demands an interoperable architecture that makes it simple to leverage new innovations from both established and new organizations as well as open source projects. Apache Iceberg, the standard open table format, and Apache Polaris (Incubating), the standard Iceberg catalog, provide the open foundation for interoperability. Dremio provides an enterprise catalog powered by Apache Polaris that can be run in any cloud or on-premises environment with:
- Open Architecture with Polaris—expands its open and governed foundation with enhancements to Dremio’s enterprise catalog powered by Apache Polaris, a fully Iceberg-native metastore. Polaris provides fine-grained access controls, robust data lineage, and deep governance capabilities across all workloads. As part of the open Iceberg ecosystem, Polaris enables seamless interoperability across engines and tools, empowering organizations to future-proof their data architecture with flexibility, transparency, and control.
Together, Dremio’s latest capabilities form a unified, open, and intelligent lakehouse that uniquely enables organizations to accelerate their AI and analytics initiatives, said the company.
"The paradox today is clear: AI demands massive amounts of high-quality data, yet teams are being asked to do more with less," said Tomer Shiran, founder of Dremio. "Our Spring 2025 release resolves this tension by eliminating the bottlenecks slowing teams down. With Autonomous Reflections and Iceberg Clustering, data teams no longer have to worry about query performance. With Apache Polaris, AI-Enabled Semantic Search, and an MCP interface, AI agents and other data consumers can instantly find what they need. We're transforming how enterprises deliver data for AI initiatives. No other platform provides this combination of speed, governance, and flexibility."
For more information about this news, visit www.dremio.com.