cv

PhD

City University of New York, The Graduate Center

Sociology

MA

Fordham University

Sociology

BA

Point Loma Nazarene University

Spanish Literature and Language

Consultant – NEPA Clean Energy Review Analysis

Clean Air Task Force

• Built a BERT+LLM pipeline to extract and classify records from 120,000+ NEPA documents, enabling CATF to analyze clean energy review timelines at scale
• Integrated Federal Register API data, targeted web scraping, and ML/regex classifiers to categorize 20,000+ projects by technology, review process, and capacity
• Scoped research questions collaboratively with CATF and delivered phased findings across six structured deliverables mapped to specific policy priorities
• Published findings via public project website and HuggingFace document browser

Postdoctoral Researcher and Lecturer

University of California, Berkeley

• Lead cross-functional research team to develop, maintain, and analyze eviction database (150M+ records) using SQL
• Applied R-based machine learning, multilevel regression, and synthetic control methods to study impact of rental assistance on eviction filings
• Coordinated with UPenn partners, managed research process, and led report writing
• Presented findings and strategic recommendations to federal housing policy stakeholders
• Taught machine learning, natural language processing, and causal inference to PhD students
• Led 32 weekly labs in R and Python libraries (tidyverse, sklearn, TensorFlow)
• Supervised 8 student projects on predictive modeling and decision-making

Data Scientist

NYC Department of Housing Preservation & Development

• Collaborated with US Census research team to establish sample survey research frame for Housing Vacancy Survey
• Addressed Housing Vacancy Survey measurement error by creating post-stratification weights
• Developed methodology with survey weights to determine where to build affordable housing, presented to non-technical stakeholders, and persuaded Commissioner to adopt methodology

Research Scientist, Professor Van Tran

City University of New York, The Graduate Center

• Built US Census dataset to study ethno-racial neighborhood integration in metro New York
• Fitted multinomial logistic regression models and interpreted marginal effects plots in R
• Wrote methodological sections, created descriptive plots in R, and managed project development on GitHub

Research Scientist, Professor Paul Attewell

City University of New York, The Graduate Center

• Developed research projects to study divergence of benefits for groups with different educational attainment
• Managed 4 large national, longitudinal, and cross-sectional datasets used in analyses
• Used data mining, HLM, OLS, and logistic regression techniques to understand the impact of educational attainment on mid-life labor market earnings

UX Researcher

CUNY, Center for Urban Research

• Designed, executed, and led focus group research to identify key labor market trends across 10 different projects
• Distilled findings into actionable insights and presented recommendations to clients

UX Researcher

Empirical Creative

• Designed and led mock jury trials (focus groups) to identify favorable juror characteristics and attitudes
• Conducted strategic research to understand user characteristics that shape case opinions and what facts change decisions
• Translated insights into actionable recommendations — trial strategy, user profiles, and visual graphics

Data Scientist – Superdiversity in Metro New York Project

Max Planck Institute

• Managed end-to-end project development of interactive Superdiversity Website and Teaching Tool
• Built and analyzed 6 cross-sectional, longitudinal databases exploring changing diversity in Metro New York
• Supervised design team and ensured project aligned with stakeholder vision

NEPA Clean Energy Environmental Review Analysis

• Trained BERT classifiers and LLM adjudication layer to extract review dates and process types from 120,000+ unstructured federal documents
• Integrated Federal Register API, web scraping, and ML/regex classifiers to categorize 20,000+ projects by technology, capacity, and review type
• Published findings via a public project website and an interactive HuggingFace document browser, making results accessible to policymakers and researchers

Voter Turnout in New York City

• Used Bayesian Improved Surname Geocoding (BISG) to impute race/ethnicity from surnames in 4.6 million NYC voter registration records
• Applied multilevel regression and poststratification (MrP) to impute educational attainment and build survey post-stratification weights
• Applied RAG techniques and fine-tuned open-source LLMs to extract insights from voter file records

SF Residential Inspection Risk

• Linked fire incidents, building violations, inspection records, and parcel tax data at the parcel level using APN identifiers
• Built composite risk scores and k-means spatial clusters to prioritize inspections for SFFD and the Department of Building Inspection
• Deployed two interactive Shiny dashboards: SFFD, DBI

Superdiversity in Metro New York

• Built 6 cross-sectional longitudinal databases and led end-to-end development of interactive website and teaching tool

Housing Literacy

• Developed online tool annotating NYC rent regulation legal documents for rent-stabilized tenants and housing advocates

New Frontiers of Integration: Convergent Pathways of Neighborhood Diversification in Metropolitan New York

RSF: The Russell Sage Foundation Journal of the Social Sciences

Kasey Zapatka, Van C. Tran

Affordable Regulation: New York City Rent Stabilization as Housing Affordability Policy

City & Community

Kasey Zapatka, Juliana de Castro Galvao

Superdiversity in Metropolitan New York: Technical Report

Max Planck Institute

Kasey Zapatka

Reordering Occupation, Race, and Place in Metropolitan New York

Springer International Publishing

Kasey Zapatka

Does Demand Lead Supply? Gentrifiers and Developers in the Sequence of Gentrification, New York City 2009–2016

Urban Studies

Kasey Zapatka