10 Ways to Make a Living as a Data Scientist - dummies

10 Ways to Make a Living as a Data Scientist

By John Paul Mueller, Luca Massaron

After reading about all the cool things you can do as a data scientist online, you might be thinking that you want a career as a data scientist. However, LinkedIn and your online network haven’t really produced what you think would be the ultimate career. Most of your options seem, well, boring.

That’s probably because you haven’t heard of all the positions that really require data science skills, even though they aren’t labeled as such! This list tells you about the amazing jobs you can get, but only if you have data science skills. The list doesn’t have specific job titles because data science is used in so many ways. Rather, what you find in this list is categories of data science–related positions.

  • Architect

    Information comes in all sorts of formats today, from a plethora of sources. Unless someone massages the data, making sense of it is impossible. Information overload becomes information blockage when you can’t even make sense of the data you’re viewing. The architect takes data from all sorts of sources and finds ways to make it useful. Most architects are computer scientists, software engineers, or people who are somehow involved in database management.


    This is usually someone who performs business-related tasks with data science, such as product management or product development. The goal is to model data to find new patterns in it and then use that model to forecast the results of business decisions. It’s akin to looking into a crystal ball to tell the future, except that this crystal ball is based on facts and statistics. In many cases, these people have some sort of engineering degree coupled with an MBA.


    While an analyst is business looking and product, businesspeople are busy looking at people. Think about someone who is not only interested in what you buy, when you buy it, and how much you’re willing to pay but also in why you buy something.

    People in this category want to recommend an incredible add-on product to you, but they want to do it in such a way that you really want to buy it, rather than feel forced into the decision. To some extent, psychology enters into this area, along with being an MBA and possibly having some science or engineering skills.


    You may wonder whether there is a place for art in data science until you start looking at the huge number of ways in which data is presented to people so that they can actually understand it. Simply spewing out graph after graph won’t get the job done.

    It takes someone with finesse and a bit of computer graphics background to get the job done right. Imagine finding someone with graphics design, computer science, and a bit of psychology all mixed into one package and you’ll understand this category.


    Almost every data scientist does some amount of application development. However, the developer focuses on making the underlying machinery of data science work. To make the big data accessible and to manage it in practical ways, you need someone who actually understands how computers work at a low level. The developer normally has a computer science degree and may have experience in other areas as well.

    Domain Expert

    Someone has to create the underlying infrastructure used for data science. The domain expert usually has strong math skills and may be a computer scientist, someone involved in finance, or someone with strong statistics skills. You get to deal with tasks such as performing natural language processing or creating financial or economic models.


    An engineer differs from a scientist in that the engineer applies known methods and techniques to data (rather than invents the methods and techniques). As an engineer, you get to see the results of the activities that the scientists around you perform, and you apply them in practical ways to real-world problems. Many of the people who work in this area are software engineers or database management specialists.


    Discovering the next step in data science is the goal of the researcher. Someone in this category likely has a PhD and is grounded in all sorts of theory that other data scientists may not even comprehend. Often the researcher comes from academia and has a creative bent. Working in this category means having an ability to speak math as fluently as someone else speaks his or her native language.


    Imagine taking art, engineering, and a real-world perspective and mixing them all together to create visualizations that anyone can understand. More important, the visualizations are concrete enough to touch and interact with in a meaningful way.

    A visualizer is someone who has great spatial skills and can perform tasks such as applying output data to maps in a way that everyone can see the results provided by data science. For example, a visualizer might help someone understand the potential future effects of global warming or the potential problems of putting someone on Mars.


    This category appears last for a good reason. Everyone knows that someone with a broad range of all the data science skills exists somewhere, but these people are incredibly rare. When working on a team, these kinds of people coordinate the efforts of everyone else and actually understand everything going on. Some people have truly amazing lives, and the unicorn is one of them.