Welcome to my site

First, a bit about my path to data science

After graduating with a BA in economics I worked for Congress' research arm, the Congressional Research Service, for four years. There I learned about how data can be impactfully applied in civic settings; the microsimulation modeling systems I developed were used to inform economic responses to the pandemic. I then left CRS to enroll in Carnegie Mellon University's MS in Public Policy and Data Analytics program to build technical skills and explore machine learning.

After a year of ML and policy coursework I joined the Internal Revenue Service as a Data Science Fellow. My team evaluated how IRS data science teams could identify and mitigate bias in their ML pipelines. In my final year of graduate studies, I worked on two client-facing ML projects with governments looking to serve their citizens better, under the tutelage of experienced practitioners.

I now work as a Senior Data Scientist with the Urban Institute on the HIPSM team, where I simulate health insurance policy changes, write papers, and incorporate machine learning into microsimulation systems.

Things I Can Do

I have worked on diverse projects requiring a wide variety of skills and policy context. Here are a few things that I like to do!

  • Code in Python, SQL, SAS, and R
  • Develop microsimulation models
  • Build machine learning pipelines
  • Design efficient databases
  • Conduct causal inference
  • Direct complex technical projects
  • Communicate findings to stakeholders
  • Research social policies

Relevant Projects

Prioritizing community outreach in Kansas with machine learning

My team is building a full ML pipeline for two counties in Kansas using the triage Python package. Our goal is to identify individuals in these communities who are most likely to experience a behavioral health crisis in the short term, medium term, and long term across varying label definitions. This multifaceted output will provide their caseworkers with a more nuanced view of a person's history and potential future.
Rayid Ghani is advising this project.

Collaborative mapping with AssetMappr

We founded this app to collaboratively map assets in deindustrialized Pennsylvanian communities. Our desktop beta was coded in Dash and hosted via AWS and our mobile beta is coded in Flutter. We collected requirements in two community pilots and are conducting two UI/UX workshops in Spring 2023 to inform the mobile app's design.
Rick Stafford is advising this project.

Improving a housing triage algorithm

I assisted researchers at the Human Computer Interaction Institute prepare followup studies on a paper which identified concerns with ML algorithms used to triage housing to unsheltered individuals. We aimed to improve these systems to better reflect the housing needs of individuals.
Hong Shen is advising this project.

Preventing ML bias at the IRS

I was selected as a Summer 2022 Civic Digital Fellow by Coding it Forward, an organization which places early career technologists in federal agencies. My data science team at the IRS examined python packages such as aif360 and aequitas and conducted a literature review on fairness in ML. From these evaluations we produced recommendations for the agency's developers.

Informing Congress with microsimulation

While working for the Congressional Research Service, I developed microsimulation pipelines using SAS, R, and the Urban Institute's Transfer Income Model. My team primarily used this pipeline to estimate the impact of various federal programs on poverty and to simulate policies that were proposed in response to COVID-19. We presented our findings to congressional staff as they considered legislative options.

Contact Me

Please feel free to reach out! For projects, jobs, or coffee.