Prepping for a Data Science interview? Ask Yourself These Questions to Stand Out
In today’s post, we hear from Alex Weiss, Engineering Manager - Machine Learning, on how to prepare for a data science interview.
It seems everyone and their grandmother wants to work in data science these days, and there are plenty of good reasons why that’s the case.
First, the salaries are attractive: recent data from O’Reilly indicates the average compensation of a data scientist in the United States is $112,000 per year (here in Western Europe, the average is also relatively high at $57,000 per year). Second, the work is challenging, rewarding and respected: By 2012, only a few years into the 21st century, Harvard Business Review had already declared data science this century’s “Sexiest Job.” Finally, career opportunities are plentiful, and only set to become more so, with IBM recently estimating that demand for data scientists will grow 28% by 2020.
While there may be a global shortage of data scientists nowadays, there’s no shortage of ways to become one. Traditional universities, online education companies, bootcamps and even potential employers themselves are all deploying courses and resources to help interested professionals train themselves in the field.
It’s clear that enthusiasm, high-quality jobs and technical training all abound in the field — so what’s stopping anyone from becoming a data scientist?\
The Knowledge Gap No One's Talking About
As someone who regularly interviews candidates for data science jobs here at GetYourGuide, there’s a troubling knowledge gap I keep encountering. I believe it’s critical we address it globally if the promise of data science as a long-term profession is to live up to the hype. It’s not a lack of understanding of the field’s principles, applications and methods; it’s a lack of understanding of the underlying business problems we’re using data science to solve.
I believe a data scientist is only as good as his or her ability to ask the right questions about data — and I don’t mean abstract, theoretical data; I mean real-life, empirical data generated by actual human behavior. So, if you’re an aspiring data scientist searching for the chance to turn the skills you’ve built into a real career in the field, consider contemplating these 4 smart business questions as you prepare for interviews:
1. Which strategic/business goal(s) is the organization currently working toward?
For data scientists, it’s tempting to say “it’s all just data to me.” The problem with that sentiment is, it’s not. It’s real information about real-world events, and as a data scientist, you’re entrusted with the responsibility to make sense of it.
You don’t have to understand every detail and intricacy of the organization or industry you’re seeking to enter, but you need to acknowledge that the data sets you’ll be handling have been generated by customers, users, colleagues and partners.
Ask yourself: What is the business model? Is the current focus of the organization growth, profitability or something else? What is its strategy for achieving those goals?
If you can answer those prompts with some degree of confidence, it’ll show that you’ve made a real effort to understand your potential new employee.
2. Which data tools and products are most relevant toward the organization’s goals? Are those tools in place?
Most organizations now know they need to take better advantage of their data, but as highlighted in another recent report from Harvard Business Review, they vary widely in their strategies and tools for actually doing so. Most haven’t made much progress at all, with industry studies showing an average of less than half of organizations’ structured data and less than 1% of its unstructured data being used toward decision-making.
Ask yourself: If building up a data science operation is like building a house, what stage of the construction process excites you most? Are you someone who likes laying the initial foundation with simple tools, doing the groundwork and putting the plumbing in place? Or do you prefer to get into the details with precision tools in a house that’s largely been built?
Data science jobs vary widely across this spectrum, and knowing which available tools the organization has (and which you’ll need) will be essential to your career success — and your career happiness.
3. How can I personally contribute the most toward the organization’s goals?
Within the data science community, there are always hype cycles around particular techniques or model classes. However, a lot of uplift can be achieved with smooth execution of standard models from your tool box. The fanciest model is not always the one that delivers the biggest win for the business.
Ask yourself: Do you want to shine through model complexity or through business success?
If the former is the case, a larger data science research team is the place for you. There, you’ll have the chance to run experiments to your heart’s content, pushing the industry forward and breaking new ground not just for your own business, but for data scientists everywhere.
But if you want to validate your work through business growth, grateful colleagues and happy customers, consider bringing your talents to a small, scrappy team that needs execution of the fundamentals.
4. Does that kind of contribution align with my professional goals? Will it give me real satisfaction in my career?
This is the most important question of all. The salaries in our field may be high, the work may be rewarding and challenging and you may always be in demand, but none of those facts can guarantee your own happiness. In fact, they can only cloud your judgment.
Ask yourself: What do you want to get out of a long-term career? Where and with whom do you want to be spending time every day? Which kinds of problems do you want to be solving? And how strongly do your personal, professional and cultural values really correlate?
If you’ve trained yourself well in data science, there’s no question you’ve got a valuable set of skills. Organizations and teams of every shape and size would be all too happy to have you. That’s all the more reason to be thorough in your due diligence when prepping for an interview — and it’s not just about brushing up on your technical chops, it’s about understanding what a potential employer stands for and needs you to accomplish.
Do that successfully, and you won’t just set yourself up for a job — you’ll set yourself up for a satisfying career in data science for years to come.
Find the original article on DATACONOMY.
Other articles from this series
Understanding Data-Driven Attribution Models
Understanding Data Products and their 4 Levels of Ownership
Laying the Foundation of our Open Source ML Platform with a Modern CI/CD Pipeline
S2E3: How the Data Products Team Drives Customer and Business Impact