Calendar of events

Hello!

We invite you to the 31st meeting of the Tricity Data Community Group! The meeting will take place at a new location – the Xebia office in Gdańsk!
At the upcoming meeting, we will talk about revolutionizing application modernization with the use of AI.

Participation in the event is, as always, free of charge, but for logistical reasons we kindly ask you to register via Meetup and arrive before 6:00 PM.

Event details
📅 Date: Wednesday, January 14
⏰ Time: 6:00 PM
📍 Location: Xebia Office, Jana z Kolna 11, Gdańsk
Tryton Business House (Building C, 2nd floor)
Google Maps: https://maps.app.goo.gl/ihwsBK14F2J5hQwp9?g_st=ic

Getting there – tips:
A multi-storey car park is located at Jana z Kolna 5. Entrance from Dyrekcyjna Street (open 24/7, paid).

Public transport to Tryton:

• trams: lines 7, 8, 10

• SKM: Gdańsk Główny station

• buses: lines 106, 108, 111, 112, 123, 130, 138, 154, 166, 167, 178, 184, 186, 189, 199, 208, 212, 289

Agenda

18:00 – 18:15 Introduction and organizational matters
18:15 – 19:15 Krzysztof Kąkol – Revolutionizing application modernization with AI

The presentation will explore practical aspects of using generative artificial intelligence to accelerate application and code modernization. Real-world project case studies will illustrate the use of GenAI for tasks such as code explanation, automated refactoring, legacy system migration, and architecture optimization – significantly reducing development time and costs. The session will also cover encountered challenges, as well as best practices and future trends in AI-supported modernization.

19:15 – 19:25 Q&A
19:25 – 19:35 Break
19:35 – 20:35 Arkadiusz Zdanowski – Triangles, Ducks, and Icebergs: a Review of Data Lakehouse Technologies for 2026

The presentation analyzes the landscape of Lakehouse technologies from a 2026 perspective. It provides an in-depth comparison of table formats (including Delta Lake, Apache Iceberg, and DuckLake), focusing on differences in metadata management, transactional capabilities, and schema evolution. The session also covers architectural concepts such as Data Mesh, Data Fabric, and the Modern Data Warehouse, clearly distinguishing market trends from real-world business applications. The presentation focuses on key trade-offs and decision-making criteria, grounded in experience from real production deployments. It helps assess when a given architecture truly makes sense and when it is merely an expensive buzzword, offering a comprehensive, vendor-neutral view of the evolution of data technologies—free from marketing hype.

20:35 – 20:45 Q&A
20:45 – 22:00 Pizza & Networking

Our speakers:
Krzysztof Kąkol
Lead Solutions Architect at Xebia. He specializes in cloud technologies – especially AWS. His areas of interest include DevOps best practices and machine learning, as well as research in speech signal processing. He is currently focused mainly on innovative applications of large language models (LLMs).

Arkadiusz Zdanowski
Cloud Data Architect at Future Processing. Designs modern, scalable cloud-based data platforms. Author and lecturer of the course “Modern Data Architectures” at the Warsaw University of Technology. A passionate experimenter — he tests new technologies in his home basement, then studies them scientifically, and finally implements them commercially.

Submit your topic!

Do you have knowledge you would like to share with others? Want to explore an interesting topic, present a case study, or try your hand as a speaker? Get in touch with us – we offer full support! Email us at trojmiasto@datacommunity.pl.

You are warmly invited!

Hi!
We invite you to the 30th meeting of the Tricity Data Community Group!

The meeting will take place at DNV’s office in Gdynia!

More information about the event – coming soon.

Participation is, as always, free of charge, but for logistical reasons, please register via Meetup and arrive before 6:00 PM.

Event details
📅 Date: Wednesday, November 19
⏰ Time: 6:00 PM
📍 Location: DNV Office, Luzycka 6E, Gdynia

Agenda
6:00 – 6:15 PM Introduction and announcements
6:15 – 7:15 PM Talk #1
7:15 – 7:25 PM Q&A
7:25 – 7:35 PM Break
7:35 – 8:35 PM Talk #2
8:35 – 8:45 PM Q&A
8:45 – 10:00 PM Pizza & Networking

Our speakers:
Details coming soon.

Submit your talk!
Do you have knowledge you’d like to share? Want to explore an interesting topic, present a case study, or try your hand as a speaker? Get in touch with us – we’ll provide full support! Write to us at trojmiasto@datacommunity.pl.

We warmly invite you!

We would like to invite you to the SQLDay Lite 2025 conference, which will take place on 16-17 October 2025.

The conference will take place in

University of Bielsko-Biała
Willowa 2
43-300 Bielsko-Biała

More details you can find on the event page.

Hi!
We invite you to the 29th meeting of the Tricity Data Community Group! At our upcoming meetup, we’ll hear talks about LLMs in the context of customer understanding and a pragmatic approach to data modeling.

The meeting will take place at Hapag-Lloyd’s office in Gdańsk!

Participation is, as always, free of charge, but for logistical reasons, please register via Meetup and arrive before 6:00 PM.

Getting there
🚆 Public transport: the nearest SKM station is Gdańsk Przymorze Uniwersytet.
🚗 Parking: the most convenient option is the Olivia Business Centre parking lots – details: Olivia Centre Parking

Event details
📅 Date: Wednesday, September 24
⏰ Time: 6:00 PM
📍 Location: Hapag-Lloyd Office, Alchemia IV – Neon, Grunwaldzka 413, Gdańsk, Oliwa – 7th floor

Agenda
6:00 – 6:15 PM Introduction and announcements

6:15 – 7:15 PM
Hubert Pietroń & Piotr Zieliński – LLM as a Judge: analyzing responses in the context of customer understanding
Can a language model generate meaningful responses in a complex business environment? Can large language models (LLMs) evaluate the quality of other models’ answers? In our talk, we’ll share the story of creating an evaluation framework for LLMs that generate customer responses within the Salesforce ecosystem. We’ll show how we leveraged user surveys to measure quality and identify shortcomings in prompts. We’ll present how we built an evaluation system comparing LLM-generated answers with those sent by human agents. We’ll also highlight key factors influencing response precision: preprocessing (identifying and removing irrelevant information from messages), prompt structure, and additional contextual data.

7:15 – 7:25 PM Q&A

7:25 – 7:35 PM Break

7:35 – 8:35 PM
Maciej Skrzos – A Pragmatic Approach to Data Modeling
The talk will cover the goals, roles, and applications of data modeling in data warehouses. We’ll walk through the most popular methods, implementation examples, and discuss what to pay attention to in the early phases of a project.

8:35 – 8:45 PM Q&A

8:45 – 10:00 PM Pizza & Networking

Our speakers

Hubert Pietroń
Data Engineer with over 5 years of experience in designing and implementing cloud-based analytical solutions. He specializes in building and optimizing ETL, automating data processing workflows, and integrating analytics with machine learning. Formerly active as a mentor and trainer in Python and SQL for both adults and children. Currently focused on exploring the potential of LLMs in analytics and decision support.

Piotr Zieliński
Graduate of Gdańsk University of Technology, Data Engineer with professional experience in machine learning projects in logistics, including demand forecasting and vessel arrival time prediction. Currently working on Customer Experience projects using LLMs.

Maciej Skrzos
Data Practice Lead at Future Mind. With nearly 15 years of experience in data engineering and leadership roles, he previously served as CTO at PlumResearch (2021–2024) and worked at EPAM Systems and Best SA Capital Group. He has built and managed data departments in finance, media, and entertainment, leading large-scale data migration and modernization programs.

Submit your talk!
Do you have knowledge you’d like to share? Want to explore an interesting topic, present a case study, or try your hand as a speaker? Get in touch with us – we’ll provide full support! Write to us at trojmiasto@datacommunity.pl.

We warmly invite you!

Hello!
We invite you to the 28th meetup of the Trójmiasto Data Community Group!
The event will take place at the Gdynia office of our sponsor – DNV!

During the meetup, we’ll delve into the details of the Delta Lake format and hear about the challenges of migrating solutions to the cloud.

Participation is, as always, free of charge, but for logistical reasons we kindly ask you to register via Meetup and arrive before 6:00 PM.
Public transport: the nearest SKM station is Gdańsk Przymorze Uniwersytet.
Most convenient parking: Olivia Business Centre parking lots – details: https://www.oliviacentre.com/o-nas/parkingi/

Event details
📅 Date: Wednesday, June 11
⏰ Time: 6:00 PM
📍 Location: DNV Office, Łużycka 6E, Gdynia

This is Building C (the one with the DNV logo on top). The event room is on the ground floor – to the left after entering the building.
Nearest SKM station: Gdynia Redłowo.
Parking: along Łużycka Street (there should be spots available at this hour).

Agenda

18:00 – 18:15 Introduction and organizational announcements
18:15 – 19:15 Tadeusz Balcer – Delta Lake: A Standard for Working with Large Sets of Structured Data
The Delta Lake format, an extension of Parquet, allows efficient work with massive structured datasets.
Originally designed for Apache Spark, it is now also supported by engines like DuckDB and Polars, which can process data efficiently on a single machine.
Using open ship movement data, Tadeusz will demonstrate how Delta Lake works, how to optimize query performance through proper configuration, and how powerful Polars can be when working with this format.

19:15 – 19:25 Q&A
19:25 – 19:35 Break
19:35 – 20:35 Grzegorz Brodny – The Slow Migration of BI: Between Quick Wins and Tech Debt
In this session, the speaker will walk attendees through a multi-year project of migrating and maintaining a BI solution originally built on Microsoft’s tech stack: SQL Server, SSIS, SSAS, and SSRS.
Over time, the solution was gradually moved to AWS cloud – in stages, using various approaches and experiencing many twists and turns.
The presentation will cover architectural changes, design decisions, technical and organizational challenges, and the impact of evolving customer requirements.
This is a technical and architectural session – a story of real-world decisions, compromises, and lessons learned. Perfect for anyone curious about what a BI migration really looks like in large organizations.

20:35 – 20:45 Q&A
20:45 – 22:00 Pizza & Networking

Our speakers:

Tadeusz Balcer
Tadeusz is a software engineer at DNV, specializing in designing algorithms for processing large datasets.

Grzegorz Brodny – Leader of the Trójmiasto branch of Data Community. Senior Data Engineer at Future Processing.
He has over 12 years of experience in building Business Intelligence solutions – from gathering requirements and writing ETLs to solution modeling.
He has mainly worked with Microsoft tools (SQL Server, SSIS, SSAS) and has been working with AWS cloud solutions for the past 3 years.
Professionally passionate about ETL optimization, columnstore indexes in SQL Server, and the Delta format. A fan of simple solutions.
He also holds a PhD in computer science in the field of Affective Computing.

Submit your topic!
Do you have knowledge you’d like to share? Want to explore an interesting subject, share a case study, or try your hand at public speaking? Get in touch with us – we offer full support! Email us at trojmiasto@datacommunity.pl.

You’re very welcome to join us!

We would like to invite you to the 17th edition of the SQLDay 2025 conference, which will be held on 12-15 May 2025.

The conference will be held in a hybrid mode – at the Centennial Hall in Wrocław and remotely.

Details are available on the official event page for more information, we hope to see you there!