In 2017, the Carnegie Mellon University (CMU) Libraries embarked on rolling out their new research repository, KiltHub (powered by Figshare for Institutions), to campus. KiltHub is intended to capture all products of research in a single repository: from papers and posters to datasets, code, and images. The CMU libraries are striving to meet researchers’ needs for library services and university support through an embedded liaison model by appointing liaisons with deep disciplinary experience who can partner with researchers and be champions for their needs. The simple, hands-on approach of Figshare including researcher self-deposit coupled with the institutional support of liaison experts provides easy and impactful sharing of research in KiltHub. An example of this model at CMU is psychology. Researchers in psychology and neuroscience have been keen early adopters of KiltHub and with the help of their liaison have deposited a variety of different types of work, at various points in the research lifecycle, and with varying goals for sharing: from data supporting publications shared to meet a publisher mandate to sets of images used as stimuli shared to be reused and cited. This has illuminated the strengths and weaknesses of serving the psychology community with KiltHub as it relates to metadata, file structure, workflow, and documentation. Meeting researchers’ needs for how and when they share their research is a critical goal of the project and this presentation will highlight what’s worked, what hasn’t, and what’s planned from the CMU, KiltHub, and Figshare perspectives. (Presented by Ana Van Gulick, CMU, and Dan Valen, Figshare). An additional version is available at: https://doi.org/10.1184/R1/5897611.v1