Sociological research has long recognized that, even in brief or routine interactions, people constantly make judgments about others’ social worlds and that these inferences have material consequences in contexts as diverse as hiring, venture capital funding, and courtship encounters. Yet it remains unclear whether people are accurate in making these interpersonal judgments and, if so, how far they can “see” into the social structure surrounding unfamiliar others. We draw on the “thin slicing” paradigm from social psychology to assess how accurate people are in making inferences about the social networks of unfamiliar others. Our data set includes over 2,100 interpersonal judgments made by 375 people about the social networks of 23 male and female targets. We find that people can make accurate judgments about unknown others’ proximate social structure—that is, the size and composition of targets’ reported contacts. They do not, however, make accurate judgments of more distal features of social structure—that is, the nature of connections among targets’ reported contacts. We also find that people’s judgment errors are broadly consistent with gender stereotypes. We conclude with a discussion of this work’s implications for research on cognition and social structure, and on the antecedents of gender inequality.