Bimetallic nanoparticles are of immense scientific and technological interest given the synergistic properties observed when two different metallic species are mixed at the nanoscale. This is particularly prevalent in catalysis, where bimetallic nanoparticles often exhibit improved catalytic activity and durability over their monometallic counterparts. Yet despite intense research efforts, little is understood regarding how to optimize bimetallic surface composition and structure synthetically using rational design principles. Recently, it has been demonstrated that peptide-enabled routes for nanoparticle synthesis result in materials with sequence-dependent catalytic properties, providing an opportunity for rational design through sequence manipulation. In this study, bimetallic PdAu nanoparticles are synthesized with a small set of peptides containing known Pd and Au binding motifs. The resulting nanoparticles were extensively characterized using high-resolution scanning transmission electron microscopy, X-ray absorption spectroscopy, and high-energy X-ray diffraction coupled to atomic pair distribution function analysis. Structural information obtained from synchrotron radiation methods was then used to generate model nanoparticle configurations using reverse Monte Carlo simulations, which illustrate sequence dependence in both surface structure and surface composition. Replica exchange with solute tempering molecular dynamics simulations were also used to predict the modes of peptide binding on monometallic surfaces, indicating that different sequences bind to the metal interfaces via different mechanisms. As a testbed reaction, electrocatalytic methanol oxidation experiments were performed, wherein differences in catalytic activity are clearly observed in materials with identical bimetallic composition. Taken together, this study indicates that peptides could be used to arrive at bimetallic surfaces with enhanced catalytic properties, which could be leveraged for rational bimetallic nanoparticle design using peptide-enabled approaches.