We have created over 60 million light curves from the TESS full frame images using a local implementation of the eleanor pipeline: a light curve for nearly every star brighter than 15th magnitude in the 26 TESS prime mission sectors. With this unprecedented set of light curves, we have developed a machine learning classification tool to identify classes of variable stars. Here I will focus on the ~300,000 eclipsing binaries identified via this machine learning classifier. I will present the overall sample, discuss our filters and sensitivity limits, and show how we have uniformly modeled them all. I will also present preliminary population results as well as highlight unique discoveries in unusual regions of parameter space.