If Databricks has its own transactional database now, what does an application built on it actually look like? That's the question this session answers, live.
Lakebase is a managed Postgres registered in Unity Catalog, with lakehouse tables synced into it and the operational data living beside the analytics under one set of permissions. That collapses a stack most teams run as three systems and two copy jobs. We'll build one application on it in the room and be honest about the edges.
What we'll build and show
- The application's own tables in Lakebase, and what a Postgres developer keeps and gives up
- Synced tables, with a lakehouse table arriving in Postgres for the app to read, and how fresh it is
- A Databricks App on top, running in the workspace under its own service principal
- An agent answering questions over the same data through AI Gateway, inheriting the user's Unity Catalog grants rather than a prompt telling it what not to say
- Genie for the questions nobody wrote a screen for
- Where the seams are, meaning what's GA, what's still in preview, and the cases where we'd keep a separate database
Agenda
- 18:00 Doors, food and drinks
- 18:30 Welcome, and what this group is for
- 18:40 Building on Lakebase, live
- 19:30 Questions, war stories, and what you'd like the next session to be
- 20:00 Close
Who it's for
Application developers, data engineers and architects in the Phoenix metro, on Databricks or considering it. If you run Postgres today and have been asked how it fits with the lakehouse, this is the session.