Join us for the next Databricks User Group Warsaw meetup, where practitioners share real-world stories from building, migrating, and running data platforms on Databricks.

Whether you’re just starting with Databricks or already running production workloads, this meetup is a great chance to learn from real experiences and connect with the local data community.

Talks

Title: What If Access Was a Product Property? Encoding multiple identity models in a single contract
Presenter: Hulya Sahip Aktas, BEC Financial Technologies
Description: Access becomes complicated when the same data product must serve different consumers with different identities. In this session, Hulya will demonstrate how to encode those access models in a single product contract and generate audience-specific enforcement.

Title: Multi-Agent Data Engineering: Claude, Codex, Local AI and Databricks with Omnigent
Presenter: Krzysztof Burejza, Databricks Resident Solution Architect
Description: What happens when instead of using a single AI coding assistant, we let multiple specialized agents collaborate on the same Data Engineering task? In this session, Krzysztof will demonstrate how Omnigent can orchestrate Claude, Codex and local models to build, review and improve Databricks solutions, and discuss where multi-agent workflows can provide real value in everyday engineering work.

Title: CI/CD on Databricks: What the Docs Don’t Tell You
Presenter: Mariusz Kujawski, Independent Cloud Data Architect & Consultant
Description: Databricks offers several approaches to CI/CD, but the documentation rarely tells you clearly which one to choose and when. Git Folders, DABs Editor, and Databricks Asset Bundles each solve a different problem, come with their own limitations, and can be the right choice depending on your team’s maturity, project scale, and where you are in your deployment journey. In this session, Mariusz will walk through the three main approaches, explain how they differ in practice, not just in theory and provide concrete criteria to help you choose the right one for your context. He will also cover what you need to build around your chosen approach, including branching strategy, environment isolation, and testing, so your CI/CD pipeline works not only in a demo, but also in production. This session is aimed at data engineers and data architects who are choosing a CI/CD approach for Databricks projects or want to revisit and improve their current setup.

Agenda

17:00 – 17:15 – Registration
17:15 – 17:30 – Kick‑off
17:30 – 18:10 – What If Access Was a Product Property? Encoding multiple identity models in a single contract
18:10 – 18:15 – Break
18:15 – 18:55 – Multi-Agent Data Engineering: Claude, Codex, Local AI and Databricks with Omnigent
18:55 – 19:00 – Break
19:00 – 19:40 – CI/CD on Databricks: What the Docs Don’t Tell You
19:40 – 21:00 – Networking

The meeting will take place at the BEC Financial Technologies office.

Who is BEC?

BEC is a financial technology company creating IT solutions for a strong group of Danish banks. For the last 60 years, BEC has enabled its customers to invest online, take out loans, and transfer money. In other words, they do everything it takes to run a bank.

Learn more at: https://www.bec.dk/poland/