Many products are designed in a modular fashion. This allows them to be adapted to the requirements of a specific usage scenario. “Configuration Design” addresses the problem of composing the optimal product configuration for a given usage scenario. We have developed a novel agent-based approach to solving the configuration design problem. The approach is based on a genetic algorithm for which the modification and evaluation operations are implemented as autonomous asynchronous agents. Specific design knowledge about the configuration design problem at hand can be included in the agents, resulting in a significant reduction of the size of the design space and of the cost of evaluating candidate designs. Furthermore, thanks to their autonomous and asynchronous nature, these agents can be easily executed distributedly on a network of workstations. The power of this novel agent-based approach has been demonstrated for the design of modular robot manipulators.