dataMinds.be

This is an In-Person Evening Session

Detailed information and registration can be found here :
https://dataminds.be/building-modern-data-applications-using-databricks-lakeflow-unlock-hidden-insights-with-vector-embeddings-in-person-u2u/

### Building Modern Data Applications using Databricks Lakeflow

Speaker: Jurgen Postelmans

Lakeflow Declarative Pipelines is a powerful Databricks framework for building reliable SQL or Python data pipelines with minimal code. It automates ETL processes, enforces data quality, and provides full pipeline visibility—whether you’re running incremental batches or real-time streams.

At the most recent Data + AI Summit, Databricks open-sourced its declarative pipelines framework—previously known as Delta Live Tables (DLT)—and contributed it to the Apache Spark open-source project. This move could also pave the way for full support in Microsoft Fabric in the future.

### Unlock Hidden Insights with Vector Embeddings

Speakers: Stijn Castelyns

Hidden relationships, contextual meanings, and subtle patterns often escape even the most advanced queries. Enter vector embeddings—an AI-driven technique that helps you analyze data in a whole new way, revealing insights that conventional tools can’t reach.

In this session, you’ll discover how to integrate vector embeddings into your existing data workflows using familiar technologies like MS SQL Server and Python with pandas. We’ll show you how to perform vector search to find similar records across large datasets and apply these techniques to a variety of real-world use cases—think customer segmentation, content classification, and anomaly detection.
Whether you’re managing databases, building apps, or analyzing data, this session equips you with practical tools to enrich your data analysis and make your solutions smarter.