As somebody who is suffering from this precise issue at another org right now, and would very much like to solve it, one glaring omission from this article is whether this worked or not!

That is, did the "inverse Conway" actually result in getting the unified observability platform shipped, and that platform actually solving the problem they initially were trying to solve? It certainly seems plausible that it COULD, but quite surprising that that's not actually stated in the article if so...

The ‘empowered full cycle developer' is the key term here. It isn't pager duty, it is an empowerment by pager.