SignificanceAlzheimer's disease affects a growing number of people, but a cure is lacking. The disease is connected to the formation of plaques in the brain, the first of which appear years before the first symptoms. Current approaches fail to stop or revert the propagation of these plaques, which are also a source of neurotoxic species in the form of oligomers. This work represents two directions toward therapeutic developments: 1) the design and production of protein libraries based on a small and stable scaffold, and 2) the realization of a screening procedure that allows for the identification of oligomer binders. The approach is successful in identifying a candidate protein that binds to oligomers and reduces the rate of plaque proliferation.