Dramatic algorithmic breakthroughs over the last ten years enabled ab initio calculations of the formation of the first stars, their impact on their environment, early metal enrichment, dwarf galaxy formation, the build up of cosmological magnetic fields and many questions relating to the interplay between galaxies and the intergalactic medium in general. With these calculations of growing complexity in hand, detailed predictions for potential observables of the very high redshift universe have become feasible. In this talk I will highlight some of interest for JWST, ALMA and the upcoming redshifted 21cm observatories. New simulations of local star forming regions explain the longevity of molecular clouds. A complete theory of star formation is needed to build confidence in any of the predictions about the so far unobserved very high redshift universe as well as our understanding of the formation of galaxies in general.