I don't suggest otherwise. It's known that Python is one of Google's three languages approved for internal use (or apparently four now that JS has got a foothold). But I'm sure I've seen reports that people are nonetheless being pushed towards Google's other two officially blessed languages, at least on production code - and in fairness, Google is a place where developer time is often cheaper than the machine time you have to spend to save developer time. But moveover it seems that, even when people in Google do want to fix their Python performance issues by making Python perform better rather than by reducing its use, they tend to want to do it in-house rather than by adopting or supporting other people's work: look at Unladen Swallow. That would be in line with the rather hubristic, command-economy approach - "let's hire all the world's smartest people, then order them to build the future" - that seems to determine Google's actions in general. Hiring GvR fits that pattern too (as well as making plenty of sense).

But I'm sure other people here are much better informed about this than I am.