Thread by @FlorisgoesF

Many young people ask me how they can become a #DataScientist, specifically in #football. Lately I have also seen a lot of posts on how to get into #DataScience in (1)50 days or so, which is a joke imo. Here is my realistic take on it. Warning: it will be closer to 1500 days. 🧵

#DataScience is an umbrella of roles & fields that require different competencies. But they all have two things in common: you have to know #Science and you have to be able to work with #data. The first requires learning to do research, the second learning to do #programming.

Go to uni and get a masters degree that at least requires some #math skills. I’m not saying you need a #PhD and 5 publications before calling yourself a #DataScientist, nor that you can’t be one without a MSc, but is helps a lot in acquiring the right competencies.

Learn #programming. #python or #R are a good start, but learning any programming language is the first step of many. Becoming a programmer is easy: print(“Hello World!”), but becoming a good one takes time and effort, and you won’t become one by just following a course online.

Practice, a lot and daily: follow courses, do coding challenges, work on your own projects, compete in competitions like @worlddataleague or @kaggle. Also: review code and have your code reviewed, it will greatly help your learning process. No one said it would be easy.

Find one or more domains that interest you, and become knowledgeable in it. #DataScience is about solving real world problems through the use of data. It’s okay to rely on domain experts to help you out, but you should have detailed understanding of the problems you help solve.

Do an #internship, preferably in a development team. It will teach you what being a #dataScientist really means. Spoiler alert: it’s about a lot more than programming jupyter notebooks. It will also allow you to learn about things like architecture, CI/CD, scalability etc.

Acquire technical skills that are essential in implementing #DataScience in practice, especially in a product. The essentials include SQL, Spark, Cloud technology (Azure / AWS) and the likes. Most of it isn’t to hard to learn, and it will get you a long way.

Work on your communication skills. If you are capable of explaining complex things to non technical people in a simple way, you can be extremely valuable as a data scientist. If you can talk to the business and development side of a company, you’ll be a key asset.

Done all that? Now you are ready to start a career as a data scientist, which probably means your learning path is just starting. The road is long and hard, and rewarding, as it should be.