The development of super-resolution (SR) microscopy in recent years has revolutionized cell biology, breaking the diffraction limit of light microscopy by order of magnitude. However, SR is currently incompatible with high-content imaging. RT-SuperES will provide a groundbreaking and affordable technology with automated SR capabilities beyond the state-of-the-art. To this end, we will generate a library of endogenously-labelled SNAP-tag fusion proteins in mouse embryonic stem cells (ESCs), and deploy a real-time decision-making module, which will continuously monitor our SNAP-tagged cells using fast fluorescence imaging, and, once a change is detected, will fix the desired cells, and switch to SR mode. By bringing together seven world-leading experts from four different countries, combining basic and applied research and industry, we propose several firsts: a) The first endogenously-labelled clone library of SNAP-tag fusion proteins; b) Utilize machine learning (ML) for real-time automated decision making, autonomously switching from fast conventional to SR imaging; c) Combine high content with SR imaging; d) Integrate novel, cutting-edge technologies, namely SR Radial Fluctuations (SRRF), NanoJ-Fluidics, Single Molecule Localization Microscopy (SMLM) and Structured Illumination Microscopy (SIM); e) Collect large scale imaging datasets of cell states in ESCs, and f) Provide cell-cycle stage-dependent nanoscale localization of selected nuclear and chromatin proteins (e.g. H3.3), during early ESC differentiation. RT-SuperES will provide the scientific community with the first-of-its-kind commercial real-time SR-highcontent imaging system, and the first library of endogenously SNAP-tagged ESC clones, which are ideal, among many other things, for SR imaging.