I am a full-time data scientist and part-time academic researcher.

Data Science
Broadly, my work is about solving administrative challenges that lie at the intersection of justice and public policy. Task-wise, I primarily do data pipelining, modeling, and (internal) model deployment, with the goal of providing insights to policy makers, research managers, and executive stakeholders at the federal level. Prior to my 2 years in the federal sector, I worked for 2.5 years in the tech and financial sectors in various data roles.

Research
My research interests lie at the intersection of computer science and law/policy.

On the computer science front, I am broadly interested in improving natural language processing (NLP) capabilities of large language models (LLMs), with the overall aim of refining machine reasoning for cybersecurity and legal applications. Specifically, I am interested in understanding how LLMs can better extract knowledge from textual structure (type embeddings, type information, knowledge base triples, etc.) and am currently investigating this architecturally via retrieval-augmented generation (RAG) and mixture-of-expert (MoE) systems.

On the law/policy side, I am doubly interested in the (1) regulatory policies about LLMs and in (2) deploying LLMs in underserved legal-political applications. From the "about LLMs" angle, I am interested in developing legal disincentives for unethical LLM deployment (e.g. cybersecurity law, legal compliance), and from the "using LLMs" angle, I am most interested in leveraging LLMs in accelerating mundane federal cases (e.g. mass adjudication, proper allocation of civil benefits and aid).

Personal
Beyond my professional and research interests, I am an avid fan of all things gamified problem-solving: board games, escape rooms, speed puzzling, and murder mysteries. I also enjoy rucking, boba, and Mongolian BBQ.