Hi @Oliver_Beier. I’m the maintainer of an n8n node called everyrow (github) that I think should be able to solve your problem out of the box. You can use it to deduplicate, rank, screen, or merge your data or just apply generic LLM calls/agents at scale (e.g. one for every data entry). I’m happy to help. And if you give me an example snippet of your data (and what the output should look like) I can create a workflow template for you.