Derek Araujo was a Fellow at Insight Data Science during Fall 2017, and now works as a Data Scientist at Schireson Associates in New York. After earning an advanced degree in law, Derek started his career at an international law firm, and later served as legal council at a think tank.
However, Derek explains that he “missed using the quantitative side of his brain” in his law career and decided to go back to his undergraduate focus and long standing interest: Physics. He went on to earn an M.A. — and later a Ph.D — in Physics at Columbia University.
During graduate school, Derek “developed an interest in data analysis and learned to love the art of coding.” He learned about Insight Data Science through his Graduate School colleagues, and believed the program could be a great fit. Data Science had become a burgeoning field in New York by that time, and he felt that a career in the field could be the perfect focus for his background and skills.
In an interview with SwitchUp, Derek talks more about how he got started in Data Science, his experience at Insight, and his advice for anyone considering a Data Science Fellowship program like Insight Data Science.
You hold a Ph.D. in Physics, and an advanced degree in both Physics and Law. What made you decide to start a career in Data Science?
Data science was a natural career fit for me. Although I practiced law for several years, I earned my baccalaureate in physics. I ultimately left legal practice to pursue a Ph.D. in physics, partly because of my longstanding interest in the subject, but also because I missed using the quantitative side of my brain. During graduate school I developed an interest in data analysis and learned to love the art of coding. By the time I graduated, data science had become a burgeoning field. A vibrant professional community had matured in and around New York City, where I very much wanted to remain after graduating from Columbia. So the match was perfect.
How did you choose to attend Insight Data Science? What was your process to research Data Science programs?
I learned about Insight through word of mouth while in graduate school. Several Columbia physics Ph.D.s ended up at Insight, and spoke very highly of the program and its job placement record. The fact that the program is tuition free made it all the more appealing. I submitted my application just after graduating, and had planned on applying to Data Science bootcamps or going directly on the job market if I wasn't admitted. Fortunately Insight accepted me soon after I interviewed with them.
What was the application process like? How did you prepare?
The application process was fairly streamlined. Candidates submit a written application and questionnaire describing their experience and interests. The Program Director selects a subset of applicants for thirty-minute Skype interviews. During the interviews the candidates are asked to talk about their academic research and review some code they have written. Admission decisions follow pretty quickly. In my case I was accepted within a day or two.
Prior to applying I devoted a good amount of time to self-studying machine learning algorithms, many of which I hadn't encountered as a physics Ph.D. student. I'm certain that served me well. In retrospect, however, I may have over-prepared for Insight's Skype interview. I spent the better part of a weekend completing a one-off project designed to showcase my machine learning and Python coding skills. My interviewer ultimately spent only ten minutes or so reviewing a small portion of my code. I suspect the interviewers use the code review to look for what would interest any employer: clean and readable code, but also oral presentation skills and an ability to explain technical concepts in an accessible way.
Tell me about the learning environment. What was the curriculum like?
There was no formal curriculum to speak of. Insight's program is a full-time, seven week fellowship. During the first three weeks of the session each Fellow selects and completes a data science project he or she will present to employers. The remaining four weeks the Fellows help one another hone their presentations, prepare for technical interviews, and make their first presentations to employers. The learning environment is highly collaborative. Fellows come to Insight with varied skill sets and experience, and are expected to help one another acquire new skills rapidly throughout the program. In the end we ended up learning a great deal from each other.
What was your biggest challenge in the program?
My biggest challenge was finding the time to complete my project while also attending the numerous break-out sessions and employer presentations that were scheduled during the day. The first portion of the program was very intense. We had barely three weeks to collect and clean our data, test our models, and bring our projects to completion. That was a tall order, considering that half the day was spent in various meetings. Although this required some long nights of coding, the end result was satisfying. It also taught us to quickly identify which aspects of our projects had the highest return on investment, and which were expendable.
Insight Data Science boosts a diverse network of fellows. Has access to this community influenced your work as a Data Scientist?
It certainly influences my work at Schireson. Many of the data scientists I work with every day are Insight alums, and we work collaboratively, just as Insight Fellows do. I also make an effort to keep in touch with the Fellows from my Insight session and previous alums, who work in just about every area in the industry. I am still fairly early into my career as a data scientist, but I can easily imagine our benefiting from one another's professional connections and technical expertise as our careers progress.
How did Insight Data Science help with your job search process?
Insight's program directors played an instrumental role in helping me to navigate the job market. They facilitated introductions to prospective employers, helped the Fellows to manage their interview schedules, held mock interview sessions, and offered valuable feedback and advice throughout. The Insight Fellows themselves obviously do the hard work of completing the projects they present to employers. But the program directors served as an important catalyst by helping Fellows to identify and connect with companies that suited their interests and talents.
Now that you work as a Data Scientist at Schireson Associates, what is your day-to-day role like?
My role at Schireson has been satisfying on multiple levels. Schireson is a strategic data science consulting firm. My work involves a blend of coding and modeling on the one hand, and client-facing work on the other. Much of my time is devoted to the bread-and-butter work one would expect as a data scientist. But at the end of the day we need to derive actionable insights from that work and communicate them clearly to clients, many of whom do not have technical backgrounds. That inevitably requires translating technical jargon into plain English. But it also involves helping clients to approach their business needs from a data analytic perspective, or helping them to identify important questions they might not have thought to ask in the first place.
How do you use the skills learned at Insight Data Science in your new role?
The firm I work for draws upon a wide variety of models and techniques, both traditional and cutting edge, depending on the needs of a particular project. I had my first hands-on experience using some of these models while at Insight. Boiling down my Insight project's results into a brief slide presentation and pitch was also a valuable experience. We are often required to do the same for clients.
What are your career goals going forward?
My overarching goal is to continue working in an environment where I am constantly learning and developing as a data scientist, while hopefully helping my colleagues to do the same. In addition to providing a large measure of personal satisfaction, it serves an important practical end. Data science is a rapidly evolving field, where stasis can lead quickly to obsolescence. Five years from now I expect to be using some techniques that are only now just maturing, or that haven't yet been invented. If I remain sufficiently agile to evolve in tandem with my chosen field, I will be happy.
What advice do you have for people interested in attending a Data Science Fellowship?
I can only speak personally to my experience at Insight, which was uniformly positive. Regardless of which program interests you, it helps to talk to current and former attendees about their experiences, their programs' strengths and weaknesses, and what preparation they found to be most helpful. In general, be prepared to work hard, and to learn quickly. Most of all, be open to learning from your colleagues and offering them help in return when they need it. This collaborative approach makes learning much more efficient, while also helping to identify any gaps in your own understanding.