TLDR: Dremio has unveiled ‘The Agentic Lakehouse,’ a new version of its data platform designed to be built for and managed by AI agents. This platform aims to simplify complex data infrastructure, accelerate AI strategies, and enhance data access and analysis through autonomous, metadata-driven capabilities, including AI Functions embedded in SQL and an AI Agent for natural language querying.
Dremio Corp. has announced the launch of ‘The Agentic Lakehouse,’ a significant evolution of its data platform, Dremio Cloud. This new offering is specifically engineered to cater to the demands of the ‘agentic era,’ where artificial intelligence systems are expected to operate autonomously, collaborating with humans to generate insights, automate operations, and continuously optimize performance.
The core innovation behind Dremio Cloud is its autonomous, metadata-driven capabilities, which enable the platform to self-manage while seamlessly powering AI agents, applications, and analytics across the enterprise. This addresses the growing challenge organizations face in managing complex data infrastructure while ensuring AI agents can effectively access and comprehend business data.
Key features of ‘The Agentic Lakehouse’ include:
AI Functions: These functions embed Large Language Models (LLMs) directly into SQL, allowing analysts and AI agents to analyze unstructured data (such as PDFs, images, and documents) with simple queries. Specific functions like AI_GENERATE, AI_COMPLETE, and AI_CLASSIFY are designed for tasks like transforming unstructured content into queryable data, intelligent text generation/summarization, and sentiment analysis/document categorization, respectively.
AI Agent: This feature transforms user interaction with data by allowing natural language queries instead of traditional SQL. The AI Agent understands conversational context, refines semantics, creates reusable views, and builds visualizations directly within Dremio Cloud. It is integrated with Dremio’s Lakehouse Catalog and governance framework, ensuring all interactions adhere to existing data policies and access controls.
Hosted MCP Server: The Dremio Model Context Protocol (MCP) Server is now fully hosted for Dremio Cloud customers, facilitating direct connections for MCP-enabled AI agents like Claude, ChatGPT, or custom models to the lakehouse. This server exposes Dremio’s semantic layer with full governance and performance optimizations.
Dremio’s Open Catalog powered by Apache Polaris: This Iceberg-native catalog provides unified metadata management, fine-grained access controls, and data lineage tracking, ensuring a governed and high-performance agentic lakehouse that can be deployed anywhere.
VS Code Extension for Dremio: A new extension integrates the lakehouse directly into the developer workspace, enabling data engineers and analysts to browse catalogs, write SQL, and visualize results within VS Code. It also includes Microsoft Copilot integration for natural language dataset discovery and SQL generation.
Sendur Sellakumar, Dremio’s Chief Executive, emphasized that Dremio Cloud aims to eliminate the manual overheads that burden data teams, boosting their productivity by up to tenfold. He stated, ‘By embedding intelligent, autonomous capabilities into a managed platform, Dremio Cloud eliminates the manual overheads that plague data teams, while providing the performance and reliability that AI agents demand.’ The company offers a free trial with $400 worth of credits for organizations to experience the platform’s capabilities.
Also Read:
- Google BigQuery Revolutionizes Data Management with AI-Powered Transformation
- Hightouch Unveils AI Agents Platform to Revolutionize Marketer Efficiency
This release signifies Dremio’s commitment to building a truly self-managing data ecosystem, where AI agents will increasingly operate as coordinated entities, further streamlining data operations and accelerating AI and analytics initiatives.
