So you want to be a data scientist? You must first become a data princess.

When I think about the relationship between the data science team (which I have been a part of) and the rest of the organization I am always reminded of Chinese foot binding. This, of course, is the tradition among Chinese aristocracy back in the day of wrapping their daughters' feet with tight bands, deforming the feet and restricting their growth. Reading Wikipedia it sounds like this didn't have much practical purpose, but I mean, let's be honest, how many times do you think someone is going to ask you to take out the trash, or go to the grocery store for some milk, or like submit a pull request, when you can barely even walk? No, no, not only do you not have time for such terrestrial tasks, you are not even capable of performing them. Indeed, you are not even capable of becoming capable. So now take note, aspiring data scientist, this is the peak of data-science enlightenment, the crowning of the data princess into a data queen: the inability to become able to perform any work that is not Data Science (extra big capital D, extra big capital S).

What's this, you already know how to submit a PR, you say? You already know a language other than R? You already have experience downloading a file from S3? Well, sadly there is no changing the past, but there is still hope: technological amnesia. And like your inability to become able, it is not sufficient to pretend that you have no experience outside of the sacred realm of data science; you must truly forget. Lest you inadvertently suggest in a morning standup that you might be able to help deploy that helm chart your team has been in desperate need of for the last three months. That way lies a decidedly unscientific little pigeon hole called "engineering".