Using the digitized cards from the David M. Rubenstein Rare Book and Manuscript Library’s old card catalogs, a team of students will explore extracting structured data to develop searchable and sortable descriptions of manuscript and archival collections. They will use textual analysis tools and natural language processing techniques to prepare an indexed digital collection of structured card catalog metadata for publication in Duke’s Digital Repository. The team will then develop ways to visualize and search this dataset based on different research topics or terms. Ultimately, the Rubenstein Library is seeking a tool that allows users to search, tag, and export description from the card catalog’s dataset. This work is a critical piece of a broader initiative within the Rubenstein Library to find and describe historically marginalized voices in our archival collections, particularly those collections documenting BIPOC history.