Luciano Strika
Aug 31, 2018 · 1 min read

Many of us find Python to be a lot more expressive, while saving intermediate states instead of creating tables may also be more comfortable.
For instance generating a row applying a non-trivial function, then aggregating that, then applying another function.
Python code could generably also be more easily maintained than some unwieldy queries, again if the logic is complicated enough.
I agree if all you’re doing is taking the mean of a column, you should probably be using something like redshift.

Of course in the end, some of this may end up being subjective!

    Luciano Strika

    Written by

    Computer Science student at Buenos Aires University, Data Scientist at MercadoLibre. I also write about Machine Learning and Data on www.datastuff.tech