We built a Ruby on Rails app to allow Chicago tenants to submit and browse complaints about landlords, incorporating data from the City of Chicago Data Portal. You can view our live app here, as well as our code on Github.

We predicted the areas in Chicago where eviction risk is highest in the next 3 years, by combining past eviction records, demographic data, crime data, and data on redlining, and then developing a machine learning pipeline. You can view our code on Github.

We examined the closure of polling locations in the United States from 2008 to 2016, and the geographic and demographic factors associated with these closures, using visualizations and regression analysis. You can view our interactive maps and read our findings at the project website, and view our code on Github.

We gathered, cleaned, and visualized data capturing the flows of money between Chicago aldermen and lobbyists, presenting it as a Shiny app. You can read about our workflow in this data.world case study, and view our datasets here. Our code is on Github.

I’m a certified instructor with Data Carpentry, an organization that conducts data management, analysis, and visualization workshops for learners with limited computational experience. Tools taught include R, Python, and SQL.

I was the original project lead on an initiative to create a code of ethics by data scientists and for data scientists, launched at Data for Democracy - an online collective of over 4,000 data scientists. In this role, I conducted a presentation and workshop on data ethics at Bloomberg’s Data for Good Exchange 2017, and moderated further discussions at the Data for Good Exchange 2018. You can read my interview with Bloomberg about the project. You can also learn more about the initiative, sign the pledge, and read this overview of how the project has evolved over the past two years.