AI/ML and Crypto are sort of generally applicable to many startups, hence you see them more. For databases, there are quite a few columnar database startups.

AI/ML and Crypto are sort of generally applicable to many startups, hence you see them more. For databases, there are quite a few columnar database startups. A bunch of these topics are basically only relevant at large scale since they're R&D intensive. I think it's easy to say that Facebook, Google, and Microsoft have big groups of people dedicated to most of these problem domains. Search their engineering blogs or look at white papers and you can find a bunch of examples.