The 'cosmic dawn,' when the Universe's first stars and galaxies formed, lies beyond the reach of even the most powerful optical and near-infrared telescopes in operation today. Even the James Webb Space Telescope (JWST), to launch in just a few years, will struggle to detect the earliest sources of light given that it is the neutral -- and thus opaque -- high redshift intergalactic medium (IGM) that hides star-forming galaxies from view, rather than any fundamental limits in detector technology. A growing body of work suggests that targeting the IGM itself, via rest-frame 21-cm emission from neutral hydrogen atoms, is *the* solution to this problem, as the 21-cm brightness encodes the temperature and ionization state of the IGM and thus the ultraviolet and X-ray emission properties of galaxies. The lunar far-side is the optimal platform for conducting such observations, as it escapes the radio-frequency interference and ionospheric complications that plague Earth-based radio observatories. In this talk, I will discuss recent work focused on better understanding the 21-cm signatures of the Universe's first galaxies, progress in signal extraction algorithms, which attempt to distinguish the 21-cm signal from foreground contaminants, and the development of physical models that unite many probes of the early Universe within a common framework. I will highlight one such probe -- the proposed Dark Ages Radio Explorer -- whose unique view from the lunar far-side will probe galaxies out to redshifts of z ~ 35, and thus provide import context for JWST measurements at redshifts of z ~ 10-15.