SendGrid was/is profitable (I've since moved on). It was a perl and python shop that turned to a go shop; a fantastic decision that greatly improved developer productivity. One team was rails and it turned out to be a trash fire before going to Go.

I'm now at an elixir place. I'm not convinced it is the right tool for what we do but it was chosen to leverage the BEAM arch. Headcount growth is the only thing in the way of being profitable at this juncture.

In short, I disagree with your thesis.