This talk introduces gRNAde, a geometric RNA design pipeline for the 3D inverse folding problem. gRNAde is a generative model operating on RNA backbone conformational ensembles to design sequences likely to fold into the backbone. On a fixed backbone re-design benchmark, deep learning based gRNAde obtains higher native sequence recovery rates (54% on average) compared to physically based Rosetta (45% on average), taking under a second to produce designs compared to the reported hours for Rosetta. We further demonstrate the utility of gRNAde for multi-state design for structurally flexible RNAs as well as zero-shot ranking of mutational fitness landscapes in a retrospective analysis of a recent RNA polymerase ribozyme structure.