AI-NATIVE BUSINESS INTELLIGENCE Volume 1 - Foundations, Architecture, and Transformation: From classical data warehousing to agentic, AI-first analyti

By VU TRI CONG MOBILUCK - CODE247.AI

Your dashboards show you the number. Who explains the why?

Every BI system built in the last twenty years was designed to answer "what happened." The AI era demands more: systems that investigate causes on their own, explain their answers, and act - without losing the discipline that made classical BI trustworthy. This book builds that bridge, end to end.

AI-Native Business Intelligence, Volume 1 is a complete, practice-first course in three parts:

PART I - Classical BI, rebuilt properly. Inmon vs. Kimball, dimensional modeling, slowly changing dimensions, ETL/ELT, data quality, MDM, and governance - the foundation your future agents will stand on, taught through one continuous retail case.

PART II - The AI-Native architecture. The six-layer reference model with the semantic layer at its heart: metrics as code, MCP as the controlled door, conversational and agentic BI, RAG and knowledge graphs, hallucination control, compile-time security, and FinOps for agent workloads. You'll learn why "wrap an LLM around the database and hope" fails - and what to design instead.

PART III - Transformation in practice. Greenfield, brownfield, and hybrid playbooks with roadmaps, risk registers, and rollback criteria - capped by a full multinational, multi-currency capstone (Truong Son Group) you can rebuild for your own organization, exercises and reference solutions included.

Inside this volume35 chapters, 60+ original architecture figures, and chapter quizzes with full answer keysReady-to-adapt templates: semantic-layer YAML, MCP server configuration, eval suites with "trap" questions, Bus Matrix, design checklistsA 90-day / 6-month / 12-month migration playbook with quantitative stop-and-rollback criteria30/60/90-day self-study paths for newcomers, BI veterans, and decision-makers

Written for data engineers, architects, analytics leaders, and CTOs who refuse to choose between AI capability and engineering discipline. No prior AI experience required; every concept is built from the ground up.

Volume 1 is the thinking book - what to build and why. Volume 2 (Design & Implementation) is the building book. They pair, but each stands alone.

• Format: Paperback
• Language: English
• ISBN: B0HDN5WGD4
• ISBN13: 9798191846798
• Release Date: August 2026
• Publisher: Independently Published
• Length: 336 Pages
• Weight: 0.99 lbs.
• Dimensions: 0.7" x 6.0" x 9.0"