I've been doing the "whole" stack: gathering and cleaning data, building the backend, researching, validating the different models...
At a first glance, I'd say that it'd be better to focus on the big picture: help define what we need to do to solve different business problems and think about ways to solve them, leaving the "implementation" stage to the other data scientist.
I think that I'm better for this kind of tasks and that also it's a better skillset to get more autonomy and leverage, but maybe I'm seeing it wrong.
What do you think? What path would you choose?