Whenever someone asks me what’s the best way to learn data science or improve as a data scientist, I always say side projects. While reading a textbook or going through a course can be helpful in establishing foundational knowledge, it’s not until you implement the ideas that you get the practical hands-on experience that translates to becoming a better data scientist.
The best way to find a side project is to scratch your own itch by solving a problem or exploring a curiosity. Interested in sports? Take a sports dataset, perform an analysis on it, and publish your findings on Medium. Better yet, you can build a web scraper to get data off of a sports website and store it in a database. This will allow you to learn the intricacies of scraping and storing data, both invaluable skills for a data scientist.
In my case, I was interested in Quantified Self, and decided on building a real-time personal dashboard for my life. At this point I had been using different apps to track my life, but none of them offered the ability to have a real-time dashboard that updated as new data was made available. By scratching my own itch, I was able to learn about API’s, how to build and deploy a flask web app on Heroku, work with different visualization libraries, design a data schema, store it in a SQL database, and many other skills that I use today in my job as a data scientist.
Next time you’re interested in skilling up in a particular area, I encourage you to frame it in the context of a side project. This way, you’ll be actively engaged as you’re learning and also have the added benefit of having something to talk about within an interview context.