Enabling Agentic Analytics with Apache Iceberg and Dremio
Published by Self-published · Tech
A practical guide to building AI-ready lakehouse systems using Apache Iceberg and Dremio for agentic analytics workflows.
Who this book is for
Data engineers, data architects, analytics leaders, AI engineers, and chief data officers.
Table of contents
- The Shift from BI to Agents
- What Is Agentic Analytics?
- The Enterprise Roadblocks
- Apache Iceberg: The Table Format for the Lakehouse
- Open Catalogs and Apache Polaris
- Query Federation: Access Data Where It Lives
- Performance Without Tuning: Reflections and Apache Arrow
- The AI Semantic Layer
- AI SQL Functions: Bringing LLMs into Your Queries
- Data Ingestion: Getting Data into Iceberg
- Governance: Securing Data for Agents
- Dremio's AI Agent
- The MCP Server and External AI Clients
- The Developer Toolkit: SDKs, APIs, and Arrow Flight
- Reference Architectures and Real-World Scenarios
- Getting Started: Your 30-Day Implementation Roadmap
Excerpt
This shift isn't theoretical. It's happening now. Companies are deploying AI agents that write SQL, answer business questions, and trigger automated workflows. But most of these agents are running on architectures that were never designed for them. The data is scattered across silos. The queries are too slow. The governance is too rigid or too loose. The business context that makes the difference between a useful insight and a hallucination is locked in people's heads, not in the platform.
