Description
90-second crash course on Databricks’ biggest launches across the Data + AI stack:
Every announcement pointed at one problem: even if an agent can reach your data, it still doesn’t understand your business. Here’s how the new stack addresses that, layer by layer:
🗄️ Agentic Data
LTAP puts operational and analytical data on a single copy in open formats, so there’s no pipeline between Lakebase and the Lakehouse to break or go stale.
🛡️ Unified Governance
Unity AI Gateway runs every agent request through one governance layer with hard spend caps and intelligent routing, so simpler tasks go to smaller, cheaper models instead of everything hitting the biggest one.
🧠 Context Engine
Genie Ontology continuously builds a live graph of your enterprise knowledge in the background, so agents get your business context automatically.
🧰 Agentic Development
Omnigent, now open source under Apache 2.0, is a meta-harness that runs your coding agents together in one place, so they share context and controls instead of sitting in separate silos.
What was your favorite launch from the Data + AI Summit 2026? Let us know in the comments!
As always, Happy Building! 😊