Abstract: An increasing number of artifacts that we do not traditionally think of as programs are being encoded in software (e.g., surveys, experiments, or legal statutes). These encodings are often ad hoc, expressed in general-purpose programming languages, with at best limited notions of correctness or even support for software testing. My research focuses on the question: how do we build tools to empower domain expert end-users --- who may not be traditional software engineers --- to leverage the methods developed for software correctness and reliability from the programming languages (PL) and software engineering (SE) communities, without requiring these users to become expert programmers themselves? I will show how the intersection of PL/SE methods and data-driven domains (e.g., social science, data science, and machine learning) is not merely the application of PL/SE methods to these domains, but also uncovers novel problems for PL/SE researchers. Finally, I will highlight how recent advances in PL theory and tooling have fueled a renewed interest in automated reasoning in legal contexts and outline future work in this domain.

Short Bio: Emma Tosch (she/her) is currently an Assistant Professor of Computer Science at University of Vermont, where she is a member of the Center for Computer Security and Privacy and director of the MaPLE lab. She received her PhD from University of Massachusetts Amherst in 2020, where she was advised by David D. Jensen and J. Eliot B. Moss. Her work has been recognized with a Distinguished Paper Award at OOPSLA 2014 and as a CACM Research Highlight in 2021. Her research focuses on applying methods from programming language design and software engineering to nontraditional domains, focusing on social science and data science applications.