Speaking at PyData Berlin 2025 about DuckDB
After speaking at PyCon & PyData DE in April, I really wanted to do it again. It was fun and helped me organize my thoughts around a topic I care about.
So why not submit something to PyData Berlin 2025?
PyData Berlin feels like the smaller sibling of PyCon & PyData DE. I’m not sure why they spun off this second conference, and I’m curious how many people will show up. All in all, it’s the third Python conference in just a few months across Berlin, Prague, and Darmstadt.
My submission
When I thought about what to present, I decided to show how I build lean data pipelines without all the cloud overhead. One of my pet peeves is the knee-jerk reaction to push everything into the cloud, usually with a big hyperscaler. That might make sense at scale, but it’s often an expensive way to start out.
Two weeks ago, my proposal was accepted. I’m excited to share this with you:
- Title: Forget the Cloud: Building Lean Batch Pipelines from TCP Streams with Python and DuckDB
- When: September 2, 16:00
- Where: Room B09 at the bcc
If you work with sensor data or TCP streams (or message brokers), this talk is for you. I’ll walk through how to turn that raw data into analytics-ready datasets using Python and DuckDB, all running on on-prem servers or minimal cloud setups.
I’ll show what has worked for me, what hasn’t, and why this leaner approach can make your life easier—especially if you’re dealing with industrial systems or running close to hardware.
Still unsure if you’re attending? Check out the full conference program.
Python Meetup Leipzig in August
Before that, I’ll also speak at the Python user group meetup in Leipzig on August 12. I’ll give a quick intro to DuckDB, do some live coding, and show why it’s a great tool for data analysis.
Details here: lpug.github.io
