United States
With over five years of experience in computational social science, my career is dedicated to the intersection of technology and law. As a data scientist, I explore the potential applications of cutting-edge methods in machine learning and natural language processing in unearthing new insights from legal data for litigation. More specifically, my work involves evaluating the ability of these methods to accurately capture the transmission of complex, nuanced legal arguments and reasoning. This experience is underpinned by my background both in social science at Duke University, as well as my legal education at the University of Pennsylvania Law School.
• Collaborate with data scientists to build and refine legal analytics models, driving development of structured knowledge systems for compliance, pay equity, and employment law domains. • Engineer and maintain scalable legal data pipelines, leveraging taxonomies, metadata frameworks, and content tagging standards to support intelligent search and content delivery. • Implement AI-driven solutions across legal content platforms, ensuring seamless workflow integration and optimizing user engagement through data-backed insights. • Conduct statistical and NLP-based analysis on federal, state, and local legal data to identify regulatory patterns, emerging trends, and actionable policy signals. • Extract, transform, and visualize complex legal datasets to support content operations, product development, and strategic editorial decisions. • Design and manage quality assurance protocols for legal data structure and governance, ensuring the accuracy, consistency, and reliability of AI-ready content. • Partner with legal editors and product teams to develop standards and validation protocols for AI-generated legal research outputs, aligning with professional and jurisdictional norms.
• Engineer prompts and content structures to optimize large language model (LLM) performance in knowledge-intensive domains, incorporating tone, clarity, and legal accuracy. • Develop and evaluate legal domain prompts and use cases for generative AI tools supporting legal research, regulatory tracking, and knowledge management workflows. • Design validation frameworks and rubrics to assess LLM-generated content for accuracy, reliability, and usefulness in client-facing knowledge management applications. • Translate complex legal language into prompt templates, ontologies, and QA workflows for AI implementation. Develop documentation, metadata guidelines, and training assets supporting AI-powered knowledge delivery.
• Use Python and SQL to conduct statistical and economic analyses of payroll, timekeeping, compensation, and other HR data for labor and employment litigation, including FLSA, California class actions, and PAGA lawsuits. • Designed and automated an OCR pipeline to extract key data from wage statements, improving accuracy and reducing processing time by 60% • Perform wage and hour analyses on meal and rest breaks, rounding, unpaid overtime, and compliance audits to support litigation and expert reports. • Research and identify new capabilities and product concepts to enhance team processes and standards.
• Developed research agenda analyzing high-dimensional data using advanced quantitative methodologies, including regression analysis, natural language processing, and social network analysis. Utilized multilevel regression modeling and multinomial logistic regression to analyze behavioral data. • Collaborate with senior researchers to design and implement research studies, and select appropriate methodologies. Communicate findings to cross-functional audiences through presentations and reports. • Trained and implemented natural language processing models to determine textual similarity between approximately 10,000 documents. • Built text classifier using TensorFlow to categorize text data from multiple sources, and to identify speakers from transcript content. • Constructed and analyzed social network of 38,261 data points using degree and eigenvector centralities in R.
• Leverage subject matter expertise in law to create, review, and optimize text prompts involving legal use cases, in order to train large language models through reinforcement learning from human feedback. • Evaluate the performance of large language models by testing, analyzing, and ranking outputs from large language models.