Skip to content
Gdańsk Bay Tech Innovation ecosystem calendar · Pomerania
/
← All events

34th meeting of Data Community Trójmiasto – Gdynia

Join to meet data enthusiasts from the Tricity, exchange experiences and draw inspiration from practitioners.

When Wednesday, 21 October 2026
Time 18:00
City Gdynia
Entry Free determined automatically

Hello! Welcome to 34. The meeting of the Tricity Data Community Group! The meeting will be held in the Gdynia office of DNV!
During the meeting, we will look at vector search and the challenges of its implementation in production environments, as well as discuss the most important news and announcements from the FabCon Barcelona 2026 conference.
Participation in the event is as always free, but for logistical reasons please register by Meetup and arrive before 18:00.

Event details:

Date: Wednesday, October 21 ⏰ Time: 18:00 Place: Office of DNV, Ul. Łużycka 6E, Gdynia
It is building E (the one with the DNV Log on top). The hall is located on the ground floor - on the left of the entrance to the building. Nearest SKM station: Gdynia Redlowo. Parking places: along Lusatia Street (this time should be empty)

Agenda

18:00 – 18:15 Introduction and Organizational Affairs
18:15 – 19:15 Radosław Szulgo, Vector search – my way or nothing! From prototype to production from Percona for MongoDB
The vector search appeared in the MongoDB Community Edition. Great! But “available” is not yet “ready for production”. I invite you to the first public presentation of Percona Search for MongoDB, during which we will go from simple PoC to the complicated reality of maintaining such a system on our own infrastructure.

What are we gonna do during the meeting?

Search evolution: We will go from a classic full text search, through semantic search and hybrid search, to RAG. All based on known data patterns from MongoDB. Impact on production reality: We'll talk about challenges that only come out on production. From problems with pipeline embeddings and their scaling to deficiencies in authentication, monitoring or automation of the "Day 2" operation. Practical solution: I will show reference architecture that solves these problems with automatic embeddeds, enterprise auth and monitoring of indices...

Content processed by AI. The English version was produced by machine translation, the “why attend” blurb was written by an AI model, the cost was determined automatically from the description, the duration was determined automatically from the description. The organiser's page is authoritative. View the original from the organiser →