Hi Dutch Databricks builders,
We are excited to announce our next in-person event, this time hosted by Xebia in their Amsterdam office.
The sessions are getting cooked and soon to be shared, book your space and set a placeholder in your agenda so that you dont miss this event.

Agenda:

17:30 - 18:25: Walk-in with drinks & food 🍺🍕
18:25 - 18:30: Introduction
18:30 - 19:15: Session I - Stop Trusting Markdown: Testing Agent Skills Across Harnesses with Omnigent
Abstract: Agent skills are increasingly shared as simple Markdown instruction bundles, but performance often varies across models and agent harnesses. This talk discusses Omnigent, Databricks’ open-source meta-harness, and explores how it can serve as a common experimentation layer across systems such as Claude Code, Codex, Cursor, and custom agents. By providing a unified way to run the same tasks, skills, tools, and evaluation workflows across different environments, Omnigent makes it practical to apply established agent and skill evaluation techniques at scale and compare results across harnesses. The broader argument is that skills should be treated as testable, versioned software artifacts with benchmarks and compatibility reports, rather than trusted solely because they worked in a demo.
Speaker: Max van den Hoven
19:15 - 19:45: Break
19:45 - 20:30: Session II - <to be aanouncend soon>
20:30 - 21:00: Networking / drinks

Don't miss this opportunity to connect, learn, and explore the incredible potential of Databricks features. Secure your spot now and be part of a community that’s shaping the future of data in the Netherlands.
Register today!