[wr639]

Interesting approach using DuckDB for the underlying CRM storage. I’m curious how well the file-system based structure scales once the dataset grows significantly.

[t0mas88]

CRM datasets don't really get that big. So DuckDB sounds like a great choice.