The All-Sky Automated Survey for Supernova (ASAS-SN, Shappee et al. 2014; Kochanek et al. 2017) continuously monitors over 100 million stars and provides an excellent resource to characterize their photometric variability (Jayasinghe et al. 2018). Using ASAS-SN V/g band light curves, we systematically search for ellipsoidal variables in the ASAS-SN V-band variability catalog. We compare simple sinusoidal fits and an analytical model for ellipsoidal modulations to identify candidate ellipsoidal variable stars. The candidates are visually reviewed to remove false positives, including eclipsing binaries. Using our analytical models, we calculate the mass ratios and Roche-lobe fillout factors for these ellipsoidal variables. This characterization allows us to identify unusual binary systems for further followup, including candidate non-interacting compact object binaries with luminous companions.