Astronomical transients are signposts of catastrophic events in space, including the most extreme stellar deaths, stellar tidal disruptions

by supermassive black holes, and mergers of compact objects. Thanks to new and improved observational facilities we can now sample

the night sky with unprecedented temporal cadence and sensitivity across the electromagnetic spectrum and beyond. This effort has led

to the discovery of new types of astronomical transients, revolutionized our understanding of phenomena that we thought we already knew,

and enabled the first insights into the physics of neutron star mergers with gravitational waves and light. In this talk I will review some very

recent developments that resulted from our capability to acquire a truly panchromatic view of transient astrophysical phenomena. I will

focus on two key areas of ignorance in the field: (i) What are the progenitors of stellar explosions and what happens in the last centuries

before death? (ii) What is the nature of the compact objects produced by these explosions and what happens when compact objects merge?

The unique combination of Discovery Power (guaranteed by planned transient surveys like LSST, combined with efforts in the realm of artificial

intelligence) and Understanding (enabled by multi-messenger observations) is what positions time-domain astrophysics for major advances in

the near future.

Ultra-faint dwarfs are the lowest mass and most dark-matter dominated systems known. The shallow potential wells make them susceptible to feedback from star formation and their low baryonic content allows us to use their stars as test-particles in the dark matter potential. The main challenge is that most stars in such dwarfs are typically very faint. Here I will give an overview of the MUSE-Faint survey, a MUSE GTO survey of 10 ultra-faint dwarfs. After introducing the survey, I will outline how the high density of stellar spectra obtainable with MUSE can be used to constrain the dark matter content and density profiles of the galaxies, while the repeated observations can be used to place constraints on the binary star content in these reionization relics. I will also discuss briefly the stellar content and the luminosity-metallicity relation at the edge of galaxy formation.