Peptide drugs have recently attracted global attention as a promising cancer therapy due to their abilities to disrupt protein-protein interaction and target intracellular “undruggable” targets. However, developing peptide drugs purely through wet-lab experiments is challenging. Computational methods, especially artificial intelligence-based approach, turns out to be a viable solution for peptide drug design because they can benefit from the know-how and feature engineering in both the small molecule field and the antibody field. In this chapter, we set out to provide a comprehensive overview for computer-aided design for cancer-targeted peptide drugs. In particular, we will cover structural modeling, virtual screening, membrane permeation, multi-objective optization, and dry-wet closed-loop development for peptide drug design.