Building on a recent one-year National Leadership Grant from the Institute of Museum and Library Services, Montana State University MSU Library will build an open source Dataset Search that promotes discovery of research datasets by using data repository APIs to automatically harvest metadata for research datasets from third-party repositories, and using human curators to create metadata records for restricted or in-progress data that is stored locally at MSU. Dataset Search complements tools like Google Dataset Search, SHARE, DataMed, and the Data Catalog Collaboration Project, with the goal of promoting discovery and access for research datasets. Objectives of the project are to support low-overhead discovery and access for research datasets especially restricted and/or in-progress health sciences data from small and mid-sized institutions; and to create analytics dashboards that allow institutions to visualize research data as a scholarly product and a driver of institutional reputation. Dataset Search will be a key contribution to community-driven, community-owned infrastructure to support findable, accessible, interoperable, and reusable health sciences research data.