The world’s leading publication for data science, AI, and ML professionals.

How Not to Become a Data Scientist

What "not to do" also matters

Photo by Karsten Winegeart on Unsplash
Photo by Karsten Winegeart on Unsplash

Data Science being one of the most popular jobs in recent years has attracted numerous people from a variety of professions. These people, including myself, spend a great deal of time to become a data scientist.

On the other side of the river, there are people who help or guide the ones trying to become a data scientist. Books, personal blogs, Youtube videos, MOOC courses are some of the tools used for helping the aspiring data scientists to cross the river.

As a result of such popularity and demand, there are a ton of resources on the topic of "how to become a data scientist". I also wrote several articles trying to make the data science Learning journey more practical and efficient.

In this article, I plan to look at the topic from a different perspective. The focus here is "how not to become a data scientist" so I will try to explain what to avoid or not to do while learning data science.

It took me about two years to get my first job as a data scientist. In this longish journey, I have done some things that I should have not. Thus, I’m kind of experienced with regards to "not to do".


Read, watch, read, watch, …

Most resources come in the form of written material or videos. In general, watching is a more preferred method than reading for learning a new skill because it takes less effort.

When I read something, I feel like my brain is more engaged with the topic compared to watching a video on the same topic.

However, both reading and watching are insufficient in terms of the required active learning. It is true especially when learning a software tool or a Programming language.

When I read or watch, I feel like I’m learning very well. However, what really happens is that I understand the topic, not learn it.

As stated in this article by Nick Dam, "changes in neural connections, which are fundamental for learning to take place in the brain, do not seem to occur when learning experiences are not active. Many research studies suggest that active engagement is a prerequisite for changes in the brain. Not surprisingly, just listening to a presentation or lecture will not lead to learning."

Watching a video on Youtube is similar to listening to a lecture. If it is not supported by practicing or hands-on experience, you are likely to fail learning.

Not to do: Do not just read and watch. Try to practice as much as possible. Hands-on experience is of crucial importance in learning data science, especially when it comes to learning a software tool or programming language.


Learn this and that and all

The scope of data science is huge. Finance, retail, image recognition, natural language processing are just some of the fields in which you might be working as a data scientist.

Although the fundamentals are the same, each field has special techniques and concepts to learn.

On top of this variety, there is a rich selection of tools in data science. You will have many options for programming languages, libraries, software tools to learn and perform typical tasks.

Let me be clear. You will fail if you try to learn them all.

What you should do instead is to narrow down your focus. For instance, one programming language and a few libraries are enough for the beginning. Instead of learning many things to a certain level, try to master one. It will definitely increase your chance of getting a job as well.

Let’s focus on a more specific example. Matplotlib, Seaborn, Altair are 3 popular data visualization libraries for Python. There is no point in learning all of them. Each has some pros and cons but it is not something that should concern you during your learning journey.

Similarly, you should try to decide which field you want to work in. For instance, if you decide on working in finance, you should master time series analysis which is not a requirement for natural language processing.

You can switch domains once you actually start working as a data scientist. However, it is best to choose a particular domain and try to master the required skills. It will make your resume or portfolio stand out and increase your chance to get a job.

Not to do: Do not try to learn every topic and tool. Try to go with the minimum at first.


Overpromise and under-deliver

I have come across many posts with a title of "learn data science in x months". The "x" is usually 3 or less. This is definitely an overpromise. It takes much more than 3 months. In fact, learning data science is a continuous process so you will keep learning even after you start working as a data scientist.

It is, of course, nice to set goals but you should not start with such overpromises. It will be a motivation-breaker after the 3 month period ends. You will realize that there is much more to learn which might cause you to think that you are not fit for being a data scientist.

What we need for learning data science is a well-structured plan with realistic milestones. How long it takes depends on your skills and knowledge, prior job experiences, and your current schedule. Please keep in mind that we are not at a sprint race.

Not to do: Do not overpromise. You can set your goals but try to be realistic. Learning data science is not a task that can be accomplished in a few months. It takes time, dedication, and effort for a longer period.


Conclusion

I have spent almost two years landing my first job as a data scientist and I’m glad I did. It was a fun but challenging journey.

If you have the dedication and time, you will reach your goal of becoming a data scientist. Avoiding what I have just mentioned in this article will help your learning journey be more efficient and smoother.

Thank you for reading. Please let me know if you have any feedback.


Related Articles