My advice is not about profit.

My "time-to-market is most important" advice (not mine, it is a very sound and well researched strategy) is about making a sustainable project:

A) most passion-only projects are abandoned at some point in order to pay for food and rent. (Yes there are few exceptions among an ocean of failed projects)

B) The other point is that even if you can run this for years before showing traction. Having users mean getting feedback, which will render some of your efforts needless.

Moreover, it was only a small part of what I said. I do think that rust doesn't make sense for other reasons, and doing a rewrite in a real project needs better reasoning otherwise you would keep rewriting forever.

Moreover2, I think you're misinterpreting the original author's intent. I'm not sure if there is interest in building AI in rust as a goal for itself. Seems to me more oriented towards the result than about the love for rust.