I think Kagi is a cool idea but I just don't have a strong usecase for it. I don't find myself googling things as much as I used to, and if anything, when I am googling things they just aren't that important. I end up going directly to certain websites instead.
> I end up going directly to certain websites instead.
Many of my Kagi searches never actually show me a Kagi interface before I get to the page that I want. Many of my search queries are putting something like "jj @gh !" in my browsers URL bar, which will navigate directly to https://github.com/jj-vcs/jj, or "nats @dh !" to navigate to https://hub.docker.com/_/nats/, etc. The ! will navigate directly to the first result, and the @gh will add site:github.com to your query.
The biggest value for me is:
- Really high SNR when searching about topics I am not a subject matter expert on.
- Really good at finding prior art on a topic while not necessarily limited to papers conferences and patents (e.g. finding obscure blog posts on a hobby project that actually is relevant prior work).
Arguably this can be achieved with frontier models but my experience is that Kagi is vastly more exhaustive and well I don't need to ask for references because they are indexed in the search itself.