Natural plant populations maintain high resistance () gene diversities that provide effective pathogen resistance; however, agricultural crops typically contain limited gene diversity, so resistance is often short-lived, as pathogens evolve rapidly to evade recognition. The () gene family of barley and wheat represents a rich source of natural genetic variation that is ideal for mining disease resistance specificities. genes encode immune receptor proteins of the nucleotide-binding leucine-rich repeat class that recognize unrelated plant pathogens by binding secreted virulence proteins termed effectors. Using DNA shuffling, we generated a variant library by recombining the barley and genes in vitro. The variant library was cloned into yeast generating approximately 4,000 independent clones and was screened for interaction with corresponding barley powdery mildew effectors AVR and AVR using a yeast two-hybrid assay. This yielded a number of MLA protein variants that interacted with AVR. Sequences of the interacting MLA variants can be clustered into three groups, all of which contain a critical residue from MLA13. Although MLA13 and MLA7 differ by 30 residues across the leucine-rich repeat domain, the replacement of leucine with serine at this position in MLA7 is necessary and sufficient for interaction with AVR in yeast and AVR-dependent immune signaling in planta. We have established a pipeline that evolves MLAs to recognize distinct pathogen effectors without the requirement for protein structural knowledge and the use of rational design. We suggest that these findings represent a step toward evolving novel recognition capabilities rapidly in vitro. [Formula: see text] Copyright © 2025 The Author(s). This is an open access article distributed under the CC BY-NC-ND 4.0 International license.