Kagi Web Index

[mertdeveci5]

Hey Kagi team - great product! I am very curious on how web indexes are built for search. Figured I would ask you guys directly as to how you went about it. Things I am curious about:

1. Do you scrape the web and crawl pages every single day?
2. If yes, isn't this way too costly of an infrastructure to run?
3. If no, I noticed you mention in the docs that you use other search engine APIs as well. How then do you make sure you remove low quality results that come from those APIs?