Her research considers the impact of algorithmic risk assessments in settings such as child welfare screening, criminal justice, and loan approvals. She is particularly interested in how techniques from causal inference and transfer learning can resolve current limitations of these systems. Her research focuses on questions of fairness.