Event

Knowledge Graph Meetup: Knowledge graphs in the era of AI

Location Helsinki or online

Date 28 May 2026

Join us on May 28 for an evening exploring how knowledge graphs are transforming AI contextualisation.

This event is hosted by Solita as part of the KG Meetup Community, and brings together experts and practitioners to share insights on the cutting edge of knowledge engineering.

Agenda (EEST times)

Date 28 May 2026

Location Solita office, Citycenter, Keskuskatu 8B, Helsinki or online

16.00	Doors open & some snacks for onsite attendees
16.30	Welcome & opening: The importance of knowledge graphs in the age of AI
16.45	From data engineering to knowledge engineering in 1-2-3 by Veronika Heimsbakk
17.30	Break
17.45	QLever: Fast, scalable, and unified knowledge graph queries by Adrian Gschwend
18.30	Project Nordic Aurora Lynx team: GraphRAG lessons learned by KAMK student team
19.00	Onsite networking
20.00	Event ends

Session highlights

From Data Engineering to Knowledge Engineering in 1-2-3 by Veronika Heimsbakk

Knowledge engineering often feels like a big leap for data engineers: new concepts, new tooling, and unfamiliar terminology.

In this talk, Veronika shows that it is actually quite a small leap to get started! By using well-known data engineering tooling, Veronika will demonstrate how you can easily acquire skills in knowledge engineering. From data frames, pipelines, and transformations–we’ll walk through how the same data can be expressed as a knowledge graph with minimal extra complexity.

The focus is practical: how to think in relationships instead of tables, how to get started quickly, and when knowledge graphs actually add value. This is not a deep dive into theory, but a hands-on introduction for data engineers who are curious about knowledge engineering and want to try it now, not someday.

We will dive into maplib, a groundbreaking Python framework for RDF, with native DataFrame support for knowledge graph construction. 

QLever: Fast, scalable, and unified knowledge graph queries by Adrian Gschwend

QLever is an open-source, Apache-licensed engine that is gaining significant attention for its approach to knowledge graph querying. It unifies graph, text, and spatial search within a single, high-performance system, with vector search capabilities on the horizon.

In this talk, Adrian will briefly introduce QLever, show how it enables fast and expressive queries across different data modalities, and explain why the RDF and Linked Data community is increasingly excited about it. Adrian will also outline how to get started and what is planned next.

Can a knowledge graph reveal Moomin family relationships and who lives in the Moominhouse? by student team Nordic Aurora Lynx

Can knowledge graph–based context improve the accuracy of AI-generated answers in a data catalog environment? In this talk, the Nordic Aurora Lynx student project team from KAMK approaches this question from a practical perspective by applying a GraphRAG approach that combines vector search with graph-based knowledge modeling.

How well can these methods infer who belongs to the Moomin family, how characters are related to each other and who lives in the Moominhouse? During the project, multiple proof-of-concept implementations and experiments were developed to explore different ways of combining data, context, and retrieval. The presentation highlights an architecture where Qdrant and Neo4j are integrated into a unified AI knowledge layer, enabling both semantic search and relational reasoning simultaneously. PoC implementations built using Microsoft RAG frameworks are presented. The focus is practical: how different architectural choices impact the AI’s ability to understand meaning, relationships, and context in data and where traditional text-based retrieval falls short.

Through these PoCs, we demonstrate that when data is modeled as relationships rather than just text, the AI’s understanding fundamentally changes.

Say hi

Meet the speakers

Veronika Heimsbakk

Veronika Heimsbakk is a Knowledge Graph Specialist with over a decade of experience as developer, architect and lead for knowledge graph projects. She is passionate in outreaching the wonders of semantic technologies and the intersection with data engineering. She was awarded Norway’s Top 50 Women in Tech in 2024. Veronika is the author of SHACL for the Practitioner.

Veronika Heimsbakk is a Knowledge Graph Specialist with over a decade of experience as developer, architect and lead for knowledge graph projects. She is passionate in outreaching the wonders of semantic technologies and the intersection with data engineering. She was awarded Norway’s Top 50 Women in Tech in 2024. Veronika is the author of SHACL for the Practitioner.
Adrian Gschwend

Adrian Gschwend is the founder of Qlevia AI, an operational knowledge platform for enterprise AI, designed to help organisations turn complex, evolving data landscapes into reliable, real-time systems. For more than a decade, Adrian has focused on making knowledge graphs scale, both technologically and in real-world applications. As an engineer, he has worked hands-on with enterprises and public institutions to solve complex data integration challenges, building systems that reflect how businesses actually operate and evolve over time. He is the founder of Zazuko and has played a key role in bringing the QLever graph engine into real-world use, helping drive its adoption in both public and enterprise environments. This experience forms the foundation of Qlevia AI, which turns fragmented data into structured operational context for trustworthy, real-time decision making.
Adrian has a strong background in open source and open data, contributing to large-scale government platforms in Switzerland and Europe, as well as working with organisations such as BMW. He also serves as co-chair of the World Wide Web Consortium RDF 1.2 Working Group. His perspective is that while AI has advanced rapidly, most organisations still operate on fragmented and disconnected data. He is focused on closing that gap by building systems where data, context and decision-making come together into a reliable operational layer that adapts with the business.

Adrian Gschwend is the founder of Qlevia AI, an operational knowledge platform for enterprise AI, designed to help organisations turn complex, evolving data landscapes into reliable, real-time systems. For more than a decade, Adrian has focused on making knowledge graphs scale, both technologically and in real-world applications. As an engineer, he has worked hands-on with enterprises and public institutions to solve complex data integration challenges, building systems that reflect how businesses actually operate and evolve over time. He is the founder of Zazuko and has played a key role in bringing the QLever graph engine into real-world use, helping drive its adoption in both public and enterprise environments. This experience forms the foundation of Qlevia AI, which turns fragmented data into structured operational context for trustworthy, real-time decision making.
Adrian has a strong background in open source and open data, contributing to large-scale government platforms in Switzerland and Europe, as well as working with organisations such as BMW. He also serves as co-chair of the World Wide Web Consortium RDF 1.2 Working Group. His perspective is that while AI has advanced rapidly, most organisations still operate on fragmented and disconnected data. He is focused on closing that gap by building systems where data, context and decision-making come together into a reliable operational layer that adapts with the business.
Anitta Rosenvall

Anitta Rosenvall has over 20 years of experience in channel graphics and information design in the film and television industry, working both as a project manager and a hands-on expert. Her work has focused especially on real-time data pipelines for live broadcasts, such as Finnish parliamentary election coverage or sports events, where delivering accurate data to screens at exactly the right moment is critical. She currently works at Solita as a Service Manager in the transportation domain, focusing on data-analytics and data catalog solutions.
Alongside her professional role, she is studying Information and Communication Technology at Kajaani University of Applied Sciences, specialising in Data to AI Engineering. Her background in information graphics, and storytelling strongly shapes how she approaches data. For her, data is a network of meanings and relationships—something that needs to be seen as well as understood. Perhaps this way of thinking resembles the world of Anitta’s favourite movie: The Godfather: individual events alone do not reveal the full picture, but once relationships, roles, and networks become visible, the story—and its underlying logic—begins to make sense in a completely different way.

Anitta Rosenvall has over 20 years of experience in channel graphics and information design in the film and television industry, working both as a project manager and a hands-on expert. Her work has focused especially on real-time data pipelines for live broadcasts, such as Finnish parliamentary election coverage or sports events, where delivering accurate data to screens at exactly the right moment is critical. She currently works at Solita as a Service Manager in the transportation domain, focusing on data-analytics and data catalog solutions.
Alongside her professional role, she is studying Information and Communication Technology at Kajaani University of Applied Sciences, specialising in Data to AI Engineering. Her background in information graphics, and storytelling strongly shapes how she approaches data. For her, data is a network of meanings and relationships—something that needs to be seen as well as understood. Perhaps this way of thinking resembles the world of Anitta’s favourite movie: The Godfather: individual events alone do not reveal the full picture, but once relationships, roles, and networks become visible, the story—and its underlying logic—begins to make sense in a completely different way.
Jyri Hiltunen

Jyri Hiltunen is a third-year student at Kajaani University of Applied Sciences. He will specialise in data management and processing, as well as the development of artificial intelligence solutions. Through collaboration with Nordic Semantics, he has expanded experience with knowledge graphs, data catalogs, and GraphRAG methods to improve contextual AI understanding.
Alongside his studies, he worked in the field of digital forensics and has over 20 years of experience in law enforcement roles. His current position as system designer has further strengthened his interest in data-driven technologies and advanced analytics.

Jyri Hiltunen is a third-year student at Kajaani University of Applied Sciences. He will specialise in data management and processing, as well as the development of artificial intelligence solutions. Through collaboration with Nordic Semantics, he has expanded experience with knowledge graphs, data catalogs, and GraphRAG methods to improve contextual AI understanding.
Alongside his studies, he worked in the field of digital forensics and has over 20 years of experience in law enforcement roles. His current position as system designer has further strengthened his interest in data-driven technologies and advanced analytics.