LCDM is empirically successful but theoretically mysterious, so it is important to stress test it in as many ways as possible. I will present searches for new fundamental physics across almost the full range of scales accessible to astrophysics, from large-scale structure (O(Gpc)) through galaxy morphology (O(kpc-Mpc)) to the structure of stars (O(npc-μpc)). I will begin by describing a method to simulate the specific region of the Universe that we live in, rather than a random one as is normally done. Such "constrained" simulations of large-scale structures allow us to question if the local Universe is special under LCDM while opening the door to novel tests of gravitational and dark matter physics. I will then describe the search for new fundamental forces using the internal structures of galaxies, which afford world-leading constraints on screened modified gravity models such as f(R). Finally, I will move below the μpc scale and ask how deviations from General Relativity would appear in the behaviour of stars. I will highlight two ways in which the study of gravity in stars could impact cosmology: it offers 1) a potential local resolution of the Hubble tension, and 2) a direct measurement of Newton's constant beyond the Milky Way.