Art?
Prior to data science, I was a professional art historian and analyst.
A common theme across my art experiences was the importance of communicating abstract concepts to a non-technical audience. This was especially challenging when I was at Christie's, as this communication often involved tens of millions of dollars of fine art. For example, when given a Picasso, I had to research that masterpiece well enough to not only predict its opening price, but also justify its evaluation to stakeholders of all kinds.
In many ways, my art experience was heavily multidisciplinary, as it drew upon both the analytical framework for pricing and the creative, open-ended domain of art history. My role was to put a numerical value on the inherently non-numerical, which over the course of my art market career, has instilled in me the values of data communicability, user-centered design, and non-linear research methodologies. (See here for if your curious about my specific art market appointments.)
While the work was exciting, I gradually realized how problematic the art market's evaluation criteria was. For starters, it remained a heavily manual process that required analysts to synthesize many texts and sales figures without the use of sophisiticated technology, which in turn left room for error and bias. Seeking a more effective approach, I volunteered to explore new tools that could enhance the industry process—an initiative that ultimately led me to discover machine learning for the first time.
I was instantly fascinated by the technology and quickly recognized that if I continued in the art market, I wouldn't have the opportunity to fully develop my technical skills. After graduating from Penn, I made the decision to turn down a return offer to Christie's and instead embarked on a journey of self-teaching in math, computer science, and data science.
Since graduating, I have been deeply committed to self-educating in this domain. I pursued self-paced programs like freeCodeCamp, studied math and computer science through textbook PDFs, and dedicated myself to numerous passion projects to bridge the gap between theory and practice. While I had to work fulltime to finance this transition, I was ultimately very fortunate to have landed some tutoring and collegiate assistant positions early on in my career, which helped me maximize the time I spent immersed in data upskilling.
Fast forward to today, I am extremely lucky to have had the opportunities I've had to leverage my data science skills towards making impact! To see examples of my experiences, please refer to my "Tailored Profiles" tab at the top of the page for more details.
I am currently pursuing data science/AI in three capacities: industry developer, academic researcher, and policy researcher. Because AI is such a complex topic at every abstraction level, I found it necessary to pursue AI from multiple angles to maximize my impact. Here's a Harvard-Tech article that stresses the importance of taking multi- and inter-disciplinary approaches to AI design.