Alumni Spotlight: Dorian Goldman – Using a Pure Math Background to Solve Problems for the New York Times

My background was in pure mathematics, focusing on variational methods and PDE in the context of mathematical physics. I did my PhD at the Courant Institute of Mathematical Sciences and the University of Paris 6 (Marie Curie).

The Data Incubator team did an incredible job of emphasizing the most important and fundamental concepts that a data scientist needs to know in his career – I know, because all of these things were confirmed in my first week at my new job.

The Data Incubator surrounded me with incredibly smart, motivated Fellows who all had similar goals and backgrounds. Being in that environment, and having the structure of the program and the demands of my project, were essential to my education.

A talented instructor is not supposed to implant information into your mind, but rather function as a tour guide in a difficult landscape of concepts of varying relevance. The Data Incubator team did an incredible job of emphasizing the most important and fundamental concepts that a data scientist needs to know in his career – I know, because all of these things were confirmed in my first week at my new job.

Learn Python, Pandas, SQL and some basic regression models to get started. Make sure to download Anaconda and learn how to use IPython Notebook. You will also want to spend time practicing basic probability, statistics and algorithm questions as you will get asked these on interviews and they are taken very seriously. These skills will not only help you pass the interview for the program, but also allow you to get the best job offer at the end of the fellowship. [Editor’s Note: for more information on preparing to be a Data Scientist, check out this previous blog post.]

Next, try to think about a project that excites you. Companies love to see someone who applied data science ideas and concepts to a project which relates to a passion or interest, as opposed to doing exploratory analysis on some random data set (although this is still great). Think about what excites you and try to imagine a cool project you could do related to that. [Editor’s Note: for more suggestions on data sources, check out our posts here and here.]

I believe the majority of my success on the job market, despite not having any real programming background, was that my project showed an attempt to solve real problems facing a demographic I knew well. In the end, companies want to make money, and they want to hire smart people like you who can recognize problems and solve them. So, prove that not only can you come up with interesting and profitable solutions to problems, but you’re really excited to do so.

Honestly no mathematical methods come to mind that are of particular use for me as a data scientist, since I worked in such a pure field in mathematics. I think what I learned during my PhD that was relevant to my current career was exploratory analysis and the ability to learn things very quickly. In particular, getting a very basic version of what you want to work, and then building on it from the ground up, learning what you need to learn as you go along.

Fundamentally, the idea of exploratory analysis as a data scientist and as a research mathematician are the same – the tools and methods are just different.