RNA is increasingly recognized as a druggable target, yet structure-based approaches for RNA still lag far behind the methods available for proteins. In this seminar, I will present our lab's effort to close that gap. I will introduce Statistical Molecular Interaction Fields (SMIFs), a physics-based framework for characterizing the interaction landscape around RNA structures, and show how this approach underpins the identification and classification of RNA binding pockets. Building on SMIFs, I will present our software for automated binding pocket prediction, designed to detect, rank, and classify candidate sites across diverse RNA folds, and discuss how it compares to existing structure-based approaches. Finally, I will move beyond static structures to explore RNA's conformational dynamics more broadly: using ensemble methods to sample the energy landscape and uncover functionally relevant states that are invisible to single-structure analysis, but that can in turn become new starting points for drug targeting. Together, these tools and concepts outline a pipeline for turning RNA structural and dynamic data into actionable opportunities for drug discovery.