Labor economists increasingly work in empirical contexts with large numbers of unit-specific parameters. These settings include a growing number of value-added studies measuring causal effects of individual units like firms, managers, neighborhoods, teachers, schools, doctors, hospitals, police officers, and judges. Empirical Bayes (EB) methods provide a powerful toolkit for value-added analysis. The EB approach leverages distributional information from the full population of units to refine predictions of value-added for each individual, leading to improved estimators and decision rules. This chapter offers an overview of EB methods in labor economics, focusing on properties that make EB useful for value-added studies and practical guidance for EB implementation. Applications to school value-added in Boston and employer-level discrimination in the US labor market illustrate the EB toolkit in action.