Michael Long

Michael Long is a research data scientist with expertise in working with open-source Large Language Models (LLMs) and using Natural Language Processing (NLP) to extract data from unstructured text. In collaboration with the National Institutes of Health (NIH) and other agencies, his knowledge helps teams find innovative ways to maximize efficiency for new and existing projects. 

Mr. Long supports a variety of internal innovation projects, such as using public, national survey tables and reports as a knowledge base for a Retrieval Augmented Generation (RAG) system. In addition, Mr. Long analyzes social factors and their connection with diabetes, supporting an established microsimulation framework to make it more efficient and more generalizable to other public health research. Outside of this work, he is conducting research on how to use NLP to extract text from high school and college transcripts to partially automate the transcription process. 

Mr. Long also previously worked as a software engineer at RTI. During that time, he contributed to a variety of data visualization dashboards, partnering with federal agencies including the U.S. Department of Justice’s Bureau of Justice Statistics and NIH on the National Crime Victimization Survey Data Dashboard (N-DASH) and the MapMECFS data sharing portal, respectively.