Aptamers incorporating chemically modified bases can achieve superior affinity and specificity compared to natural aptamers, but their discovery remains a labor-intensive, low-throughput task.Here we describe the 'non-natural aptamer array' (N2A2) system, which enables fully automated, high-throughput screening of base-modified aptamers using a minimally modified Illumina MiSeq instrument. We demonstrate the capability to screen multiple different base modifications to identify the optimal choice for high-affinity target binding. We further use N2A2 to generate aptamers that specifically recognize protein posttranslational modifications, and which maintain strong target affinity in serum. Finally, we demonstrate comprehensive profiling of single-and double-base aptamer mutations to rapidly identify key sequence motifs responsible for binding activity in a single run. N2A2 requires only minor mechanical modifications to the MiSeq and a software suite for automation that we have made freely available. As such, we believe our platform offers a broadly accessible and user-friendly tool for generating custom reagents on-demand.