I will present the characterization and initial results from the QUEST-La Silla AGN variability survey. This is an effort to obtain well sampled optical light curves in five extragalactic fields with unique multi-wavelength observations, using the wide field QUEST camera on the ESO-Schmidt telescope. The survey uses a broadband filter called the Q-band, which is similar to the union of the g and r SDSS filters. We obtain good photometry to a limiting magnitude equivalent to r ∼ 20.5. I will present our results on variability-based AGN selection. We have used Support Vector Machine (SVM) algorithm to classify objects as either AGN or non-AGN using the variability features and the optical colors. We have identified ∼ 1500 AGN candidates in the five fields based on variability and color data only.