I’m responsible for advanced analytics at Ursa Health, which covers any method we use to discover information in the data, such as visualizations, statistical models, and machine learning algorithms. The Ursa Health Platform has several built-in applications to facilitate this discovery process, and part of my job is to help create and maintain these tools.
In addition, our clients often have questions that require custom approaches. In these cases, my job is to map the client question to the appropriate methods, perform the analysis, and then communicate the technical results back within the applied context. Because the initial work often reveals interesting findings that then lead to additional questions and insights, the process ends up being highly collaborative and iterative.
I studied economics as an undergraduate and then worked in the energy industry as a data scientist, although this was before it was called “data science.” I went to graduate school to study cognitive science and wrote my dissertation on models of human causal learning. Through this experience, I discovered a passion for the literature on statistical and machine learning. I’m very happy to be working in healthcare, where the work is both intellectually stimulating and provides opportunities to have a positive impact on patient care.
Ursa Health had a predictive modeling project right as I was wrapping up my Ph.D. During my participation on that project, I came to really appreciate the Ursa team. Everyone at the company is both highly capable and well-grounded, which makes for a truly collaborative environment.
You often hear the “80/20 rule” in data science, where 80 percent of the time is spent on data wrangling (e.g., organizing, cleaning) and only 20 percent is spent on actual analysis. That rule has been flipped while I’ve been at Ursa Health, so I get to spend most of my time doing analysis, which is the fun part. I think this is a testament to the efficacy of the Ursa Health Platform for delivering data that is immediately ready to be put to work.
I’m a big fan of Boston terriers—they look like little monsters!