Engineered proteins offer the potential to solve many problems in biomedicine, energy, and materials science, but creating designs that succeed is difficult in practice. A significant aspect of this challenge is the complex coupling between protein sequence and 3D structure, with the task of finding a viable design often referred to as the inverse protein folding problem. In this work, we introduce a conditional generative model for protein sequences given 3D structures based on graph representations. Our approach efficiently captures the complex dependencies in proteins by focusing on those that are long-range in sequence but local in 3D space. This graph-based approach improves in both speed and reliability over conventional and other neural network-based approaches, and takes a step toward rapid and targeted biomolecular design with the aid of deep generative models.