What You Need to Know to Transition to a Career in Data Science

Article hero image

Demand for developers with specialized skills is on the rise across the board, and companies are particularly interested in filling data science roles. A recent IBM study projects that the number of work opportunities in data science will increase 15% in the coming years, adding 364,000 openings by 2020. It also predicts that demand for data scientists and engineers will grow by 39% in the next five years. Why has data science become so important to businesses? “The explosion of data science is motivated by businesses capturing orders of magnitude more data about their customers, processes, products, and services than in the past, everything from website user behavior to turbine engine diagnostics,” says Chris Albon, data scientist and the creator of Machine Learning Flashcards. More than ever before, companies have the ability to gather and store vast amounts of information. Data scientists are responsible for converting that data into insights that can inform decisions, reduce risk, and uncover surprises. Albon says, “Making this data useful requires specialized knowledge and skills that sit somewhere between statistics, computer science, and software engineering.” As more companies rely on data to make better decisions, they’re hiring for positions in data science to help them get there. If you’re a developer interested in transitioning into a career in data science, here are a couple things to know about the role.

You’ll Learn New Things

When transitioning from a role as a developer to a position focused on data, your existing computer science and software engineering skills will be highly valuable. However, data science roles also require sophisticated statistical knowledge. If you want to work in data science, you need to hone your modeling and machine learning chops. Strong SQL skills are table stakes for data scientists and data engineers. These roles also require competence in statistical modeling and machine learning tools, such as the Python data science ecosystem that includes NumPy and scikit-learn or the rich statistical tooling of R (my own data science toolkit of choice!).

You Can Start Right Now

The most effective way to develop data science skills may be to begin where you are right now. It can be a challenge to get a new job with the title “data scientist” when you don’t have experience or a PhD in statistics, but you can still implement statistical techniques and try out machine learning methods in your job as a developer right now. Steph Locke, data science consultant/trainer at Locke Data, emphasizes the data workforce shortage that she sees. “Data science is growing because a lot of organisations have tons of data and not enough people to make all the worthwhile decisions off the back of it. We get things wrong, we don't have time, or we just don't make the optimum choice.” What are the benefits of you developing your data science skills right where you are? You can make a difference in your own company by using rigorous statistical methods, and companies can fill their data science needs by finding and training people with data-oriented mindsets from within their own organizations. If this doesn’t quite work out that way for you, then no problem! You have experience and projects to talk about when pursuing your next opportunity.

You Need Creativity and Communication

It’s important to know that data science work involves a high level of creativity and communication. Data scientists are highly technical, but also often work closely with marketing, sales, and operations. The ability to communicate effectively with people from diverse backgrounds is important. This communication happens through data visualization, writing, and speaking. In my own job, I communicate with people from software developers to product managers to sales people, and I need all of them to understand, for example, a model I have built: what it means, how to interpret its results, and how certain we can be in its predictions. Being an effective data scientist also requires creative, strategic thinking. Doing data science isn’t just about optimizing a model’s predictive power. In fact, grinding away to eke out slight improvements in accuracy would often be a bad way for me to spend my time! Instead, data scientists creatively consider which problems to work on and how best to get to actionable insights. If you’re considering starting a career in data science, now is a great time to make the transition. Equipped with these thoughts, you now know a bit more about what to expect in a data-focused role. If you’re interested in seeing what kinds of data science jobs are out there, check out Stack Overflow Jobs, where you can find hundreds of listings for data scientists right now.

Login with your stackoverflow.com account to take part in the discussion.