From Monolith to Mesh: How to Model Data in the Age of Data Products and AI Agents

Data products aim to make data easily accessible and reusable by their consumers. For this reason, data modeling is not a mere implementation detail, but a core aspect of product design. While schema-on-read has deprioritized modeling in recent years, the rise of data product–centric architectures and AI Agents is bringing it back into focus. Established techniques like dimensional modeling, data vault, and the unified star schema offer useful foundations, but most were built for centralized systems. In this presentation, we will explore how to adapt these techniques to distributed, modular architectures to create AI-ready data products that are easy to use, reuse, and compose. This presentation aims to be a synthesis of Chapter 10 of the book "Managing Data as a Product", which is dedicated to data modeling in distributed environments. We will cover distributed physical modeling, centralized and distributed dimensional modeling, data vault modeling, unified star schema modeling, and managing the physical model lifecycle.