We investigate the evaporation dynamics of a sessile droplet of ethanol-water binary mixtures of different compositions laden with alumina nanoparticles and compare with the no-loading condition at different substrate temperatures. Shadowgraphy and infrared imaging methods are used, and the experimental images are postprocessed using a machine learning technique. We found that the loading and no-loading cases display distinct wetting and contact angle dynamics. Although the wetting diameter of a droplet decreases monotonically in the absence of loading, the droplet with 0.6 wt % nanoparticle loading remains pinned for the majority of its lifetime. The temporal variation of the normalized droplet volume in the no-loading case has two distinct slopes, with ethanol and water phases dominating the early and late stages of evaporation, respectively. The normalized droplet volume with 0.6 wt % loading displays a nearly linear behavior because of the increase in the heat transfer rate. Our results from infrared imaging reveal that a nanofluid droplet displays far richer thermal patterns than a droplet without nanoparticle loading. In nanoparticle-laden droplets, the pinning effect, as well as the resulting thermo-capillary and thermo-solutal convection, causes more intense internal mixing and a faster evaporation rate. Finally, a theoretical model is also developed that satisfactorily predicts the evaporation dynamics of binary nanofluid droplets.