.png)
Internet is a scary place. Especially for people who are trying something new or changing careers. There is as much wrong information as right and as much discouragement as encouragement on there. This is specifically true for data science. What it takes to become a data scientist is a somewhat controversial topic and many people have very strong opinions on it.
Specifically, there is a group of people who seem to give off only negative messages about becoming a data scientist. I call them the data science elitists. They tend to believe that to become a data scientist you need a PhD, you need to be a maths expert and you need to know every tech tool ever made related to data, on top of your flawless computer science knowledge. Needless to say, I don't agree with them. In this article, I want to tell you why not to take data science elitists seriously and how you can have your shield up against negativity on your way to becoming a data scientist.
First of all, you shouldn’t take people seriously when they give you a long list of technological skills. A data scientist is not a developer. Yes, there are tech tools you need to learn but many of them are not required, just good to have.
Moreover, every company has its own tech stack, that is, the set of tools they use. Chances are, every company you apply to will be using a different tech stack and you can't learn all of them. Especially when you’re just starting out, knowing how to use speciality tools is not a priority. I talked about exactly this with Samantha Zeitlin (a hiring senior data scientist), on the 4th episode of the podcast. She mentioned, in her experience too, you learn the tools on the job.
You shouldn’t take people seriously when they tell you, you cannot, or it’s too late. For some reason, some people tend to give off the vibe that you either have these skills now or you’re toast. This is my face when I read comments like that:
Everyone has to be a beginner at some point. No need making someone feel bad for not having this or that skill yet. If you want to become a data scientist, you can, period. That’s what I think. It might take you more or less time depending on your background. The time it takes might also depend on your financial requirements, and if you can work on it full time or not but that’s all. This is not to say being a data scientist is for everyone. You should understand the requirements and implications of the job before diving deep into learning it of course. But I’ll talk about that in a second.
You should take people’s advice with a grain of salt. Unless someone is a professor of “making others a data scientist”, I doubt they are being objective about their job. If someone got a PhD in math and then became a data scientist, it is not surprising to hear them say you need a PhD or you need advanced math skills. People are trying to be helpful but it’s hard for them to be objective. So instead of taking one person’s advice to heart, try to read what other people have to say about the same topic. Maybe the So you want to be a data scientist podcast can help you there. *wink wink*
Of course, I cannot tell you every way you should keep an eye out for the data science elitists. You're going to have to spot misinformation (or unhelpful information) for yourself. What I can do is to show you how you can learn to only take what’s useful to you.
These are the things I go over in my free mini-course Data Science Kick-starter. You can sign-up for it at the bottom of this page. Back to our topic.
First and foremost, you need to be at a point where you’re confident of the type of data career you want to pursue. If all you have is a vague explanation of what you think you want to do, you will be easily manipulated by what you read or hear. When someone says, you need to be a good developer to become a data scientist, it might be hard for you to understand what that exactly means. Learning about the data pipeline, what type of development is required for it and which type of data professional deals with what parts of it will be helpful for you there.
You need to have a good understanding of the discipline. Artificial intelligence, machine learning, natural language procession, deep learning, cognitive solutions, computer vision, robotics, any many more are things you probably see all around the internet, mentioned closely with data science. But how are they related to each other? How much of this do you need to know to be able to call yourself a data scientist? It might seem hard to grasp at first but if you have a good overall picture of the discipline data science operates in, you’re going to be less susceptible to distractions and feel more sure of your path.
Lastly, you need an understanding of the core requirements and fundamental skills for data science. You can always build on top of the basic skills and specialize in a certain area but not without a strong foundation. Though, even after you focus on the fundamental skills, it is helpful to know in which order to tackle them and learn them. I call it the skills matrix. Knowing the required skills, in which order to learn them and being clear on how they contribute to your learning is something that will put you ahead of your competition.
In the Data Science Kick-starter mini-course, I help you choose and be confident about a data profession, explain the discipline of AI and how data science and other buzzwords fit in the picture, explain the data science pipeline and talk about the must-have skills needed to become a data scientist. It is a free-of-charge course and you can sign up for it using the form below.
After all is said and done, it’s in your hands to keep it efficient and smart while learning data science. Equip yourself with the correct perspective and nothing can stand in your way.