The National Database for Autism Research (NDAR) is a human-subject data repository on tens of thousands of research participants. Approved researchers have access to an unprecedented volume of item-level clinical, genomic, and imaging data. Data are shared quickly using both a common data standard and innovative tools for experiment definition, which provide the level of detail needed for efficient use of the repository. As described, early adopters have used it to conduct secondary data analysis. Now, with an ever-increasing volume of research data being made available, and new methods for data query, data download, and computation in place, this initiative is becoming vital to those interested in scientific discovery in autism or is being used as a model by other research communities.