When your product catalog grows to tens of thousands of SKUs, each with hundreds or even thousands of unique properties, traditional search breaks down fast.

In this session, Florian Lenz will walk you through a real-world implementation where he built a highly efficient, intelligent product search experience on top of a PIM system using Azure AI Search.

He'll take you through the key architectural decisions he faced:
- how to model complex, property-rich product data into a search index without sacrificing performance
- how to handle the sheer variety of more than 1,000 different product attributes across categories
- how to design a search and filter experience that actually serves shoppers, not just the data structure behind the scenes.

You'll get an honest look at what he built, why he made the trade-offs he did, and how he tackled challenges like index design at scale, faceted navigation across heterogeneous product attributes, relevance tuning, and keeping the index in sync with an ever-changing PIM source of truth.

If you're building or planning a search solution on Azure, this session will give you a practical blueprint you won't find in the documentation.