Marko Lohert

Marko Lohert is a senior software developer with 20 years of professional experience. He has received Zagreb County ZTK Annual award for results achieved in programming, and robotics. During his career Marko has worked on web and desktop applications.

He has presented at a number of conferences in Croatia, and Bosnia and Herzegovina. Also, he has talked at meetups in Austria, Croatia, and Bosnia and Herzegovina. In his free time he volunteers. One area where Marko volunteers is Mentoring Byte, a program for mentoring (new) ICT professionals. Another area where Marko volunteers is STEM where he teaches children programming, electronics (Arduino), and robotics.

Talk: Comparison of RAG Architectures

When we need to enhance an LLM we can use RAG (Retrieval-Augmented Generation), but which RAG architecture should we choose for our specific use case? This talk introduces different RAG architectures and compares their strengths and weaknesses.

The traditional RAG approach is to use a vector database to store embeddings. Another method is Graph RAG, which uses a knowledge graph to map relationships between entities, and this way we can get better results. We can also select Agentic RAG, which uses autonomous AI agents. There’s also Multi-Agent RAG. And the list of RAG architectures does not stop here.

Get your free ticket here https://www.entrio.hr/en/event/dorscluc-2026-28348 and see you at the DORS/CLUC 31!