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

My Honest Advice for Someone Who Wants to Become a Data Scientist

What I wish someone would tell me before studying data science

Photo by GRAY on Unsplash
Photo by GRAY on Unsplash

So, I have received many messages asking for advice and tips on breaking into Data Science. Even though I have written several articles detailing the roadmaps and processes I would follow, I think sometimes it’s good to hear the brutal truth. That’s why, in this article, I want to give you my complete, honest advice for those of you wanting to become a data scientist.

Note: You can also watch the video version of this blog post.

Start NOW

I often get asked if I should use platform A or platform B. Which course, out of course X and course Y, is better?

I have said it many times before and will say it again, IT DOESN’T MATTER.

Any large course provider or highly rated course will teach you the same things, particularly at the beginner and entry levels. The time you spend deciding what course to choose is much better spent actually doing the work, as that will get you ahead and push you towards your goal.

Similarly, people often ask me what project they should do. My response is always the same: Choose anything you like the look of and go for it. Again, wasting time choosing the "right" project is futile because such a thing doesn’t exist. You are much better off just picking something and working on it, and you can always change or choose another one you prefer later on.

As the famous quote goes

You can’t steer a stationary ship

Or as the successful entrepreneur Noah Kagan puts it:

Now, not how

If you are interested, I have written a complete list of my favourite data science and Machine Learning resources.

📖 My Best Data Science Resources

It’s Going to Require Work

A few years ago, you could get an online certification and probably land a job within a few months, but that’s not the reality anymore.

Online certifications help you learn the content in data science and machine learning, which is very valuable. However, they rarely help you get hired nowadays, especially in our tough job market.

You must do a lot of extra work like marketing yourself, creating a portfolio and doing several quality projects to get that edge over other applicants. You have to work harder now, and that’s just the brutal truth.

I know "work harder" is far from helpful, and I am not promoting a "grindset" mentality, but you now have to put in more time than people used as the supply of data scientists has increased exponentially.

If you have a full-time job, you will probably need to study in the evenings after work or on weekends. If you are a student, you will need to take courses alongside your university work.

Sacrifices need to be made, and you will probably need to do this for several months before you feel ready enough to apply for jobs, maybe even a year in some cases.

I understand this may not be feasible for everyone, and I get it. However, if data science is something you really want to do, sacrificing a year or so for a career you will enjoy for decades seems like a good decision, but of course, I am biased!

If you want a full breakdown of how to make your data science application stand out, check out my previous article.

How to Make Your Data Science Application Stand Out

Start Using Data NOW

Even though you may have to study outside of hours, try to use data as much as possible in your current role. Not only will this benefit your learning, but it will also help you decide if a career in data science is right for you.

You can start with simple tools like Excel and incorporate them more into your daily tasks. By experimenting with complex functions, you can gradually build your confidence. You can even explore VBA with Excel, adding another layer of proficiency to your skill set and becoming more familiar with programming.

If possible, you can try to do a bit of SQL and Python, as you will then be doing a data role already, making it much easier to find a job in the space later. Not to mention, your proficiency with these tools will increase exponentially.

Every company can use data in some way, as the volume of data is growing exponentially over time. Therefore, there is much opportunity to use this data at every organisation, you just have to get a bit creative!

You may even find that the value you generate with your data skills at your current company is so good that they permanently employ you as a data scientist. I have heard of several cases where this has happened.

Don’t Worry About AI

The media and non-experts have done a great job of hyping AI, which has led to some scaremongering. Don’t get me wrong; the technology is still amazing and definitely helps in certain areas. I even use it daily to boost my productivity.

Several people have messaged me questioning whether they should still try to get into the data science field because of AI. In my opinion, as it stands, it definitely won’t replace data scientists, so you shouldn’t worry about this.

NeetCode has done a great video explaining how current AI is incapable of replacing programmers.

Sure, AI can write code and do a multitude of other things. However, its definitely not at the skill level of data scientists, particularly when it comes to any form of mathematical reasoning. Try getting ChatGPT to prove Fermat’s last theorem; it just won’t do it correctly.

Even the so-called "software engineer killer" Devin is not as good as the creators initially marketed it. Many companies are trying to boost their investment by hyping AI, and their results are often overexaggerated.

When I was building a website, GPT even struggled with simple HTML and CSS, which you can argue is its bread and butter.

There are numerous examples of these Large Language Models failing or not being very useful for many tasks. This is because they have knowledge but no natural intelligence, as the famous computer scientist Yann LeCun put it:

And a 4-year-old child has seen 50 times more data than today’s biggest LLM:

Overall, don’t worry about AI; this is no excuse for you wanting to become a data scientist.

Getting a Job Is Just the Beginning

Like anything, the real work starts when you get the first role. Getting your first job is difficult, but once you are in, there will be many mental battles along your journey.

The career on the outside may seem all sunshine and rainbows, but it’s not always like this. You can expect"

  • Lifelong learning -> You will need consistently up skill in you career.
  • Constant imposter syndrome -> You will never feel like you know.
  • Feelings of burnout -> There is always something new to learn, so it’s mental exhausting to keep up with everything.
  • Constant changes in the field -> New advancements and technologies come out every year for you to keep on top of them.
  • Ambiguous job definition -> Data science is still a new profession, so it’s not exactly clear what the roles and responsibilities are.

And that’s just the tip of the iceberg.

These are universal experiences that every data scientist is likely to confront at some point in their career. If these challenges seem daunting, it may indicate that this career path isn’t for you.

This is not to discourage you but to be completely transparent about what you will be in for.

Check out my previous posts for a full breakdown of the realities of being a data scientist.

Navigating the Realities of Being A Data Scientist

7 Regrets From My First Year As A Data Scientist

Summary & Further Thoughts

I hope this article didn’t come across as too "tough love," but I just wanted to be to the point with the advice some budding data scientists need to hear. I hope it has helped you in some way and clarified your journey into data science.

Another Thing!

I have a free newsletter, Dishing the Data, where I share weekly tips and advice as a practising data scientist. Plus, when you subscribe, you will get my FREE data science resume and Learn AI roadmap!!

Dishing The Data | Egor Howell | Substack

Connect With Me!


Related Articles