Synthetic biology offers the possibility of engineering a wide variety of information processing circuits in vivo. Non-coding RNA has recently emerged as a substrate of choice for engineering new biological circuits due to its small size and programmability. We propose a new strategy for the novel generation of synthetic regulatory networks based on RNA that will allow integrating complex gene circuits with small sequences of a few hundred base pairs. For this, we developed a general methodology based on biophysical principles to automatically design RNA components able to transduce environmental signals into RNA, process the RNA through cascades of RNA interactions to eventually regulate gene expression. We are able to endow RNAs with complex features previously only found in transcription factors: non-linearity, feedback, signal transduction, multimeric riboregulation, “switchable” RNA cascades and anti-termination RNA switches. Furthermore, we are able to engineer regulatory RNAs able to self-circularize after undergoing a maturation step. Our RNAs are different to any known non-coding sequence and their predicted behavior is validated in E. coli at the population and single-cell levels. The RNA networks can be interfaced to either the translation or the transcription (CRISPRi) machinery to regulate target genes. The latter allows us to place our RNA networks in any living system where CRISPR is known to work. We also show that we can sense the transcript levels of a target gene in vivo, which we use to determine the cell-cycle phase in eukaryotic cells. Our RNAs can form complex interaction networks to provide novel synthetic gene networks working in prokaryotic and eukaryotic systems.