Molecular docking are widely used computational technics that allows studying structure-based interactions complexes between biological objects at the molecular scale. We developed AMIDE, a framework that allows performing inverse virtual screening to carry out a large-scale chemical ligand docking over a large dataset of proteins. Molecular inverse docking has several applications in the field of drug research like identifying potential side effects of existing or new drugs, or to help choosing the less harmful treatment for a disease. We developed different strategies to distribute the docking procedure, as a way to efficiently exploit the computational performance of multi-core and multi-machine (cluster) environments. This tool has been validated on well-known experimental structures through 24 protein-ligand complexes extracted from the Kellenberger's set. Its ability to reproduce experimentally determined structures and binding affinities highlighted that AMIDE allows performing better exploration than existing blind docking methods.