There is currently a wealth of disease-associated proteins that are "undruggable": inside the cell and inaccessible to large biologics but lacking small mol. binding sites. Stapled peptide therapeutics, short chains of amino acids where two are covalently crosslinked, are an emerging therapeutic framework that fill this gap. However, their discovery is reliant on solid-phase synthesis, limiting throughput and evaluation.We employ a novel technique known as Stabilized Peptide Engineering by E. coli Display (SPEED) to rapidly characterize ∼109 stapled peptide therapeutics. In this technique, libraries of DNA encoding peptides are designed by mining linear peptide data. Stapled peptides are formed directly on the cell surface by incorporation of azide residues in place of methionine and reaction via copper catalyzed click chem. Finally, cells are sorted for given mol. properties by incubating with fluorescently labeled target protein.In this work, we demonstrate SPEED's utility by applying it to the Bcl-2 protein family, important regulators of cell death. First, we utilized on-cell staple scanning to identify staple locations that maintain target binding. Interestingly, some of these locations resulted in changes in specificity for different Bcl-2 proteins. Next, incorporation of Next Generation Sequencing enables quant. measurement of affinity and specificity, streamlining downstream anal. Finally, peptide stability was optimized by treatment with peptide-degrading enzymes before fluorescent antigen treatment.To our knowledge, this work represents the first high-throughput study of stapled peptide Bcl-2 inhibitors. Peptides with these improved drug-like properties may promise better selective treatment for cancer with in vitro and in vivo validation of efficacy. This approach provides a framework to discover potent stapled peptide inhibitors towards other important protein targets.