Monoclonal antibodies capable of recognizing antigens with high affinity and specificity represent a wellestablished class of biological agents. Since the development of hybridoma technology in 1975, advances in recombinant DNA technologies and computational and biophysical methods have allowed us to develop a better understanding of the relationships between antibody sequence, structure, and function. These advances enable us to manipulate antibody sequences with the goal of improving upon, or creating new biological or biophysical properties. In this review we will focus on recent successes in using structure-guided computational methods to design antibodies and antibody-like molecules with optimized affinity and specificity to antigen and for enhancing protein stability.