Data acquisition is outpacing data analysis in every scientific discipline. As a result, the process of *doing* science is fundamentally changing: Discoveries are increasingly made by querying large public databases rather than conducting primary experiments (consider GenBank and SDSS, for example). This shift mandates collaborations with computer scientists and technology experts, especially those in the database community.As part of that community, my group builds systems that facilitate this new kind of data-intensive science. Our recent work focuses on a database-as-a-service systems targeting scientists (SQLShare and Myria). In the past, I've worked on a query language for "gridded" datasets such as those produced by environmental simulations.