A new measurement strategy called compressed sensing has recently gotten much attention. The basic tenants of this approach are that one can obtain more information with fewer measurements than "classically" allowed (for example, sub-Nyquist sampling of a signal) when one exploits prior knowledge about the system. This has lead to a revolution in sensor design that allows for application-specific optimization of measurement schemes. I describe here my recent work in compressive molecular imagining using x-rays. In contrast to previous imaging schemes that are limited to projection images only (standard radiograph), require specific sample preparation (x-ray diffraction machines), or require multiple acquisitions to generate volumetric data (computation tomography), my scheme allows for snapshot volumetric molecular imaging of an object. In order to achieve these results, we leverage recent improvements in x-ray detectors, computational tools, and compressed sensing theory.