Peptides exhibit some of the features of ideal drugs, most notably high bioactivity and specificity, however their therapeutic development is generally complicated by poor bioavailability and pharmacokinetics. Notably, some small microbially-derived peptides (e.g. cyclosporine A) overcome these limitations through the inclusion of non-canonical structural features such as macrocyclisation and backbone N-methylation. Display methodologies (phage display, mRNA display, etc) are well established techniques for the isolation of peptides with high affinities to a target of interest, but are largely restricted to the use of canonical residues, greatly limiting their use for the synthesis and screening of peptide libraries with structural similarity to drug-like microbial natural products. However, genetic code reprogramming techniques circumvent this restriction, allowing the display screening of such non-canonical peptide libraries, and thereby facilitating the identification of drug-like molecules.1,2 Compounds identified through this approach exhibit nanomolar binding affinities and (often) inhibitory constants with respect to their cognate protein targets and can inhibit protein-protein interactions. The discovery of several such molecules will be discussed, focusing on recent technical developments that allow the identification of relatively hydrophobic, low molecular weight compounds comprised mostly of non-canonical residues, which are ideal candidates for hit-to-lead development.