I’m a Data Scientist, Of Course, I…

5 Powerful Habits That Have Made/ Are Still Making Me a Better Data Scientist

Anjolaoluwa Ajayi

Published in

Level Up Coding

5 min readApr 18, 2024

Courtesy of the famous TikTok Trend “I’m an xyz, of course, I zyx.”

We are what we repeatedly do. Excellence, then, is not an act, but a habit.
— Aristotle

Cambridge Dictionary defines a habit as something you do often and regularly, sometimes without knowing you are doing it.

And I think that’s what the TikTok trend is all about: People of various identities talking about their habits and other things that make up who they are.

I’m basically jumping on the bandwagon (but in a slightly different way).

I’ve picked many habits over time from great data scientists and people who generally inspire me.

These habits improve my data science workflow and skills and help me stay ahead of the curve.

Now, I’m bringing these same habits that continue to make me a better data scientist to you in the form of a (seemingly overdone) TikTok trend (I just think it’ll be fun, you decide!) ><

TL, DR; I hope you enjoy reading this and even find some new rewarding habits to add to your data science routine :)

I’m a Data Scientist, of course, I…

… take advantage of existing solutions when put on a new task.

When I want to begin a new project, after concluding on the objectives, success metrics, etc my next instinct is always to search and see how people have done similar things in the past.

A simple Google search will go a long way, but my favorite places to look are Kaggle, GitHub, DataCamp, and YouTube.

The point of this is not to just copy and paste code or ditch your own critical thinking.

No, doing this just helps me get a better sense of direction, exposes me to new techniques, and inspires new ideas within me.

It’s like a headstart, so I know I’m not starting from scratch; just finetuning, improving, or completing solutions I need (that coincidentally already exist).

… fill my bookmark bar with frameworks’ documentation links

I mostly work on my web browser (Google Colab specifically) so it only makes sense that I bookmark the documentation of the frameworks and libraries I use regularly for easy reference.

Some of these libraries include Pandas, Matplotlib, Scikit-Learn, Seaborn, and Tensorflow.

Having them on stand-by comes in handy when I

Want to customize some parameters
Use a new functionality I’m not familiar with
Am stuck on an error AI assistants can’t explain
Need code examples of how to use certain functions.

… regularly dump code snippets, research papers, and cheat sheets in a well-organized folder

I was a writer before I became a techie. So this habit is just something I borrowed from my writing endeavors.

When reading or watching anything, I’d always dump sentences, phrases, figurative expressions, or even full paragraphs that I thought were creative or cool in a file.

I had a warehouse-sized collection of these. Not the most organized, but it was easy to find my way around with the Microsoft Word ‘find’ feature.

So usually, when I’m feeling artsy, I try to imitate the writing style/ use verbatim (for phrases and figurative expressions) in my own writing.

The same works for me in data science. I have a folder, and then 3 subfolders in the folder.

TL;DR

First folder: a repository of research papers.

Second folder: data science/ ML books and cheat sheets.

Third folder: code snippets from my previous works and other people’s works.

Focusing more on the third folder… whenever I read a case study, project implementation, and so on, whether on Medium, Kaggle, or whatnot, I always painstakingly store the interesting or novel code snippets in a file.

I also store code snippets of the different data science processes I’ve done in the past.

I give these code snippets very descriptive names, such as EDA for high dimensionality data, Customer segmentation — clustering with GMM, recommendation system-collaborative filtering, time series analysis with LSTM, etc

Many a time, when starting a new project, I always fall back on these code snippets. And then tweak them to fit the new task.

This greatly helps my workflow and saves me a lot of time.

… read Data Science newsletters/ blogs every day

Reading is a habit that improves a person irrespective of what their career or calling is.

So, no surprise that reading things related to my field every day makes me a better data scientist every day.

My day is never complete without reading one of my favorite newsletters: Daily Dose of Data Science.

Avi Chawla delivers production-level optimization tips and intuitive explanations of data science and all it entails to my inbox daily via his Daily Dose of Data Science newsletter.

This has had an amazing effect on my data science journey over the past year. After indulging in the newsletter, I always feel more confident, like “Yeah, I’m a 1% more informed data scientist than I was the previous day.”

I can’t count how many times the random data science tips I’ve read in this newsletter have saved me in my day-to-day work. It’s really top tier.

If you want to subscribe to Daily Dose of Data Science, click here.

Medium Data Science blogs and publications can also be very great sources of data science knowledge. I read at least 20 Data Science articles on Medium weekly. This habit helps me cover skill gaps and stay up to date, be more confident as a professional, and provides me with fresh perspectives.

… write about data science and other things related to it

Maybe this habit is partly because of my writing background but I always preach ‘writing about what you know’ not to impose writing on people but because it’s genuinely helped me grow as a data scientist.

If you can’t explain it simply, you don’t understand it well enough.
— Albert Einstein

If you’re able to explain what you know in a way that even beginners can easily understand, then it means you understand it well enough… which ultimately makes you an expert in the topic.

So, writing about data science and things related to it: