Architecture question

[jorgemgomes]

Hi team,

How are you doing?
When someone types a question:

• How does GenBI “know where to look” and which engine to use? I mean, when a user asks a natural-language question, how does GenBI determine which database/engine to query (e.g., Trino vs Redshift vs SQL Server)?

• How does GenBI handle cases where multiple engines could answer the question?

I’m just trying to understand how GenBI avoids generating SQL for the wrong engine.

Thanks