In late 2017, the University of Pittsburgh, Health Sciences Library System, became a partner of the Data Catalog Collaboration Project (DCCP). Using the open-source code made available by the collaboration, we implemented a data catalog at our institution, branded as the Pitt Data Catalog. Our mission is to promote open science, facilitate the discovery of hard-to-find biomedical research data, and reduce the perceived barriers and burdens of data sharing. Our definition of data is broad and includes research outputs such as analysis code and 3-D printable models as well as traditionally thought of datasets. To date our catalog includes 71 curated metadata records with fields such as description, controlled keywords, software used, and related publications. The catalog is not a data repository so each record also includes dataset access information that varies from a pointer to a public repository, an author's e-mail address, or a link to a data-access application. This lightning talk will highlight some successes and challenges we have had locally and as a collaboration. These include developing flexible metadata, conducting outreach to diverse populations, and proving value to stakeholders across campus.