A five-step process to increase the chances of landing a data science job.
The most frequently asked question is, “How do I get into data science?” Many people want to pursue a career in data science but are unsure how to get started.
And I’m not going to lie: it’s a difficult move.
You’re up against businesses and recruiters who are always searching for applicants with experience, when you’re looking for a way to start developing your own. So, what are your options?
Though I cannot guarantee your success, I can provide you with a step-by-step guide to help you increase your chances of landing your first data science work. It’s a method I used when I first started out in the industry, and it includes key steps that I’ve seen others use to take their first steps. I hope that by following these five steps, you can feel like you are making real progress toward your goal of securing your first data science work.
Are you looking for a specific type of role?
One of the difficulties with data science is that it isn’t really a single sector, but rather a set of occupations. When you first start out, think about what kind of job you want and what kind of business you want to work for. I see three roles that are fairly common:
Analysts are primarily concerned with turning data into actionable information. Dashboards, monitoring, statistics, and data mining are often used to accomplish this.
A data scientist typically develops predictive models that are intended to add value to a business. A data scientist, for example, may develop a model to predict which customers would leave.
Research Scientist — dedicated to advancing the state-of-the-art. Normally, this will include doing research and presenting papers at conferences.
Spend some time learning more about these functions, as well as any others you see in the data science industry. If you can narrow down what position you want to pursue, finding a job will become much easier. For example, if you want to be a research scientist, you’ll need a different set of skills and experiences than, say, an analyst.
Different positions can also be more difficult to obtain. Analyst roles are the best for me, followed by data science, and finally research scientist roles. This is mostly due to the high demand for these positions. In today’s world, almost every organisation employs analysts in some capacity, but few employ actual research scientists.
Once you’ve decided on the type of job you want, think about the type of business you’d like to work for. Companies may be thought of along a variety of axes. Scale, sector, and geographic location are just a few examples. Since various types of businesses employ data positions in different ways, the type of business matters. Large companies tend to have a very stringent hiring process that most applicants would go through, while smaller companies are more likely to recruit based on a recommendation or relation.
The first step is to consider the type of job you want and the type of business where you want to work. Spend some time thinking about it and writing down your ideas. Keep it close at hand because it will have an effect on the next phase.
Begin to build your network.
Now that you know what kind of job you want and where you want to live, it’s time to start networking.
Have you heard of this adage?
“Twenty years ago, the perfect time to plant a tree was. Now is the second best time.”
When it comes to networking, the same idea applies.
The worst time to begin networking is when you are in desperate need of it. It’s probably too late at that point, but if you find yourself in that situation, the only thing you can do now is start networking. However, keep in mind that, like a newly planted tree, a newly formed network is delicate, and don’t expect it to bear much fruit. Instead, concentrate on expanding and nurturing your network to the best of your ability, and wait for it to bear fruit once it has matured.
But how do you expand your network?
My best advice is to start interacting with everyone you encounter right away. And the best way to connect with others is to really want to learn more about them. Tell your cashier, “What was the best part of your day?” the next time you’re in the shop. That is a topic that causes people to pause and consider their responses. Then, depending on the answer, it may or may not lead to a more in-depth discussion. That’s fine. You’re learning how to interact with others and gaining a better understanding of those around you.
Take this concept and apply it to LinkedIn. Go join in the discussion the next time you see anything important on an article!
Extend this concept to the next meet-up or meeting. Get to know the people around you by engaging with them and learning more about them.
What you’ll notice is that you’ll get to meet a lot of new people! Many of these interactions will fade away after the first conversation, but others will develop deeper roots and grow into friendships. Continue to provide about them. Talk with them on a regular basis and see how they’re doing, what new projects they’re working on, and if there’s something you can do to support.
The best networking advice I can offer is to concentrate on giving. Helping others not only feels good, but it also establishes a solid base so that if you do need assistance, the network will be there for you. It won’t be uncomfortable either! You won’t be approaching strangers for a job, but rather friends who want to help.
It takes time to construct a network like this, so get started now! Today, find a way to interact with someone different!
Better Understand Your Goal
Okay, now it’s time to collect some information so you can better grasp the road ahead of you.
The work boards should be your first stop. Investigate the top three job-posting sites for the type of position you listed in phase 1. That could be as easy as using large sites like LinkedIn or Glassdoor, or more specialized sites like the YC job board. I’d also recommend looking at individual business job sites for your ideal employers.
Examine these work listings and start keeping track of what you find.
Is Python, R, SQL, or Spark mentioned?
What about reinforcement learning or deep learning?
Perhaps any engineering concepts will come up, such as unit testing or continuous deployment.
Your aim is to begin to comprehend what core skills are required for your ideal job. This process will simply confirm what you already knew, depending on how well you understand the job you want. However, you might come across some surprising popular themes.
Then, on LinkedIn, look for people who already have the job you want.
What abilities do they seem to possess? What are their educational and professional histories, as well as their previous experiences?
Combine what you’ve learned with the information you’ve gathered from work listings. You should now have a spreadsheet with a long list of skills and experiences, as well as a count next to each one indicating how often it came up in your study. Sort the table so that the skills that are used the most are at the top. Since data science is such a broad profession, having a well-organized list of skills is important. It’s nearly impossible to anticipate every potential interview subject. We’ll use your list to help us concentrate our efforts in the next move.
Now is the time to solidify your foundation — choose the top five skills on your list. These are the abilities you should have a strong grasp of.
This is also the point at which you must be truthful with yourself about where you are and what your goals are. If you find that you know very little about the top five skills required but still need a job right now, you can refocus your current efforts on a more attainable job while you prepare for your dream position.
However, don’t be too cynical about your current situation. If you’re having trouble assessing your abilities, see if you can reach out to people in roles similar to the one you want to get input on your strengths and weaknesses across your growing network.
Work your way down to your top five and spend as much time as you need to feel comfortable with them.
Here are some tools for some of the most common data science topics:
Machine Learning with Scikit-Learn and TensorFlow Linear Algebra: A Practical Guide Think Stats Analytics: Python for Data Analysis, Gilbert Strang Statistics: Doing Bayesian Data Analysis; Introduction to Statistical Learning
Pick the next top five skills on your list and make sure you have a basic understanding of them once you’ve mastered the first five.
You should feel confident that you have a good understanding of the core skills needed for the job you want at this stage. When you begin interviewing, this cornerstone will be invaluable.
Unfortunately, being a successful data scientist does not always imply passing an interview. You must clearly prepare for interview questions.
To assist you in your preparation, use Google:
“How to Achieve Success in a Data Science Interview” and “Common Data Science Interview Questions” are two resources to help you prepare for your data science interview.
These searches will provide you with a wealth of tools for practising actual interview questions. Every day, practise questions, focusing on the ones you believe are most important to the top skills you’ve identified.
Use Around the Board
Don’t count on the recruitment process to always get it right because it’s broken. You have no influence about whether an organisation has a poor hiring process, but you do have control over the number of positions you apply for. As a result, make certain you apply for anything.
I believe it is important to maintain the mentality that, for the most part, it is not your responsibility to screen yourself out prior to applying. Obviously, you should not apply if you are not fit for the job. But if you’re on the verge of anything, go for it! Especially if you think you’d be a good match but lack a few years of experience compared to what the role “needs.” It’s not unusual for organisations to discover during the hiring process that some of their criteria were too stringent, and to loosen them up a bit based on the applicants they receive.
Here are some pointers if you’re having trouble getting interviews:
For referrals, reach out to your network.
Make your resume stand out! Make sure you use a standard resume format that recruiters can easily search. To avoid being filtered out by bots, explicitly mention the skills required in the work posting. Also, make sure to emphasize the work opportunities that are most important to the job.
Have a friend or colleague look at your resume and provide you with constructive criticism.
Since this is the beginning of your work funnel, make sure to optimise it. You’ll easily run out of work to apply for or the resources to keep applying if you have to apply to hundreds of jobs to get a few interviews.
I’d suggest applying to around 20 positions and then waiting a few weeks to see what the response rate is. If it’s too low, make some changes to your resume and try again for another 20. Continue to iterate in the hopes of increasing the acceptance rate.
That concludes the discussion. My 5-step process will ideally assist you in obtaining a data science role. It is undeniably an operation, and one that is far from easy. So, stick with it and try to build on each step of the job funnel on a regular basis.