For the midterm elections, I built scrapers to power live election night predictions, visualized model results, contributed to data pipelines, analyzed FEC data and prototyped and refined a find-your-district personalization feature.
Designed and developed a scrollytelling map of the properties bought by a Reno developer for this ProPublica investigative project.
For STAT, built a mobile app for iOS and Android to showcase their stories and better engage mobile readers. The app lets readers follow their favorite topics and reporters, save stories for later reading and login with their STAT+ account.
For fun, built this interactive graphic to showcase data collected by an open source Big Local News project to scrape layoff notices from all 50 states. Contributed open source scrapers to the project for several states as a volunteer.
To show the state of presidential money, crunched the data behind this graphics story designed and developed by Julia Wolfe of FiveThirtyEight. The analysis used Python code and relied on a real-time SQL database of campaign finance information.
Built a tool that helped newsrooms compile databases of Trump officials' financial histories. This was used to report a daily story by the New York Times, to build ProPublica's Trump Town news app and to report an award-winning national radio story with Reveal from the Center for Investigative Reporting.
Released on April Fools' Day 2020, this browser extension replaced Gmail’s Smart Replies with lines of poetry. It was enjoyed by many, including Gmail project manager Paul Lambert, who wrote, “This is amazing. I wish we had thought of it!” It was mentioned on Lifehacker and The Verge.
Analyzed a universe of PACs that raised from small-dollar donors and spent mostly on themselves. Zeroed into one telemarketer's charitable solicitation operation with a reporter. Built graphics to tell the story.
This was an ambitious multi-part series on the places and people with little political clout. Designed and developed the presentation, as well as overseeing data analysis and graphics production.
Created a quick Serverless API and Observable notebook for this interactive knowledge test which asked users to point to Ukraine after a cabinet secretary asked an NPR anchor if she could. It received a million attempts, only 300,000 of which were from bots.
Designed the presentation for this striking story on a preventable death and what it says about workplace safety enforcement.
Helped a reporting team collect, track and input data from nearly every state legislator's financial disclosure in the country, then built an app using DocumentCloud to search and view the disclosures.
Built a map using Mapbox.js to display millions of ship locations collected by satellites, found a cabinet secretary's shipping firm docked in Iran at the height of sanctions, designed graphics for the resulting story.