Type Ia supernovae (SNe Ia) are superb distance indicators and are used to map the expansion history of the Universe. In the last millennium, astronomers used observations of SNe Ia to find that the Universe's expansion is currently accelerating. This discovery resulted in the Nobel Prize for Physics in 2011. Since this initial discovery, we have used SNe Ia to loosely constrain the nature of "dark energy," which drives the accelerated expansion. To improve our dark energy constraints beyond our current basic understanding, we must design new and better SN surveys and techniques. I will present the Foundation Supernova survey, a new high-fidelity, low-redshift (z < 0.1) SN survey started in 2015 that will replace the current heterogenous low-redshift sample and reduce the (currently) largest uncertainties for SN cosmology. I will describe the survey, our implementation, and first results. I will also discuss the next major leap in SN cosmology, WFIRST, which will launch in about a decade. I will show the first simulations of the WFIRST SN survey and make some suggestions for how to further improve this mission. With the combination of the Foundation survey, WFIRST, and new physical understanding, SNe Ia will remain a premier cosmological probe, continuing the legacy started decades earlier.