I don't have any advice to offer, but I'll give you my passing impression: it strikes me as the epitome of "ML as a solution looking for a problem". The reason

I don't have any advice to offer, but I'll give you my passing impression: it strikes me as the epitome of "ML as a solution looking for a problem". The reason people aren't getting excited is because most people are excited by outcomes, not by tools and technologies. I doubt that many people in marketing or real estate can visualize what ML could possibly do for them. That's where you need to focus imo - the killer application that no one (in some specific domain) can do without.