Solves a need, but Apple (if you use both a Mac and iOS) and Pushbullet can do similar functionality way better. Still, I'm glad people are building these sorts of things.
The question doesn't really have a well defined answer. It would vary tremendously by the organization and project they're working on. Also depends a lot on their level. There are senior engineers who code probably <25% of their time, but add value by helping other engineers, planning, architecting, removing obstacles, etc.
The article's good. However, the comment about experts not getting to the endgame as often as non-experts is likely wrong. I don't have the data either, but I'm fairly experienced at chess, and in my experience high level games get to an endgame moderately often, but amateur games almost always end before an endgame. Amateurs typically don't even know how to play an endgame, because they rarely get to it so they don't realize why they should learn. I know a lot of amateurs focus a lot on the opening because they see themselves winning or losing in the opening, when that's more due to tactical blunders than anything essential to the opening.
It'd be interesting to see some statistics for this.