Trust me,

I'm a Data Scientist

Thoughts by Sarah Diot-Girard

Aug 2, 2022

My favorite talk of PyData Berlin / PyCon DE 2019

Choosing which talks you attend at a conference is always a bit of gambling. You might end up listening to a talk not intended for you — either rehashing basics you know by heart, or so complex you’re lost after the speaker introduction slide. You might have hoped this talk to be about the theory behind GANs, and it ends up focusing on practical implementation issues — or vice versa.

There are talks that you’ll take about enthusiastically to your coworkers and friends, reminding fondly their key take-aways or their exceptional delivery.

And, from time to time, there is a talk you go back to every six months, and that you keep sharing with everyone because “if you have some time, please watch it, it’s really great”.

Mock hell is one of those talks, joining Gary Bernhardt’s talk on onion architecture and Daniele Procida’s insipiring take on writing documentation on the list that talks I’ll keep sharing with everyone all the time, in particular more junior developpers.

Now that I’m talking about it, maybe I’ll share soon a list of ressources around software engineering and data science that I use as references.

Mock Hell, by Edwin Jung, is a great talk about the too-often misunderstood principles of OOP, testing and in the end, architecture. If you struggle with writing a clean test suite, then you probably should change how your code is organised.