As a scientist, I take seriously my responsibility to communicate findings in a clear, friendly, and accurate way. My results have to speak to every member of a cross-functional team; if they don't understand it, I take is as my own fault.
To this end, I use data visualization every day in my work, and at every stage of a project, typically culminating in some kind of report, interactive dashboard, or academic publication. In my experience, good visualizations can light up a room, spark participation from shy people, and advance a group's thinking when it is stuck.
Over the last year, I have developed an informal data visualization course with 6 hours of instruction, designed for clinical researchers' needs. I have delivered the course three times to various audiences at Stanford, both in person and remotely, and I am still working on the content.
The course is based on a one-hour talk on data visualization for general audiences, which I developed in 2014 while I was at Optum. I have now given the talk to many audiences: yearly to all staff at Ariadne Labs, various other venues in the Harvard Medical School system and hospitals, and every fall semester in a computer science course at Brown University. Other one-off presentations have been through word of mouth invitations.
I also have given invited talks at a variety of venues on differences in how people comprehend data, and data visualization during the COVID-19 pandemic. I enjoy public speaking and teaching, especially in person rather than on camera.
I enjoy visualizing data in my spare time, typically with unusual or historic datasets. For example I used R Shiny to build an interactive visualization of all of the questions in Shakespeare's Hamlet (link at right).
Other personal projects have drawn from medieval court and church records, the Domesday Book, the complex family life of Mia Farrow, US age-specific fertility rates since 1920, an ancient Roman calendar, and mathematical concepts such as Simpson's paradox and the central limit theorem. I have fun.