We describe an evolutionary algorithm for the design of an imaging triple-étalon Fabry-Perót interferometer (MFPI), which gives a solution to the multidimensional minimization process through a stochastic search method. The interactions between design variables (the étalon reflectances, interétalon ghost attenuator transmittances, and spacing ratios) are complex, resulting in a fitness landscape that is pitted with local optima. Traditional least-squares and gradient descent algorithms are not useful in such a situation. Instead, we employ a method called evolution strategies in which several preliminary designs are randomly generated subject to constraints. These designs are combined in pairs to produce offspring designs. The offspring population is mutated randomly, and only the fittest designs of the combined population are passed to the next iteration of the evolutionary process. We discuss the evolution strategies method itself, as well as its application to the specific problem of the design of an incoherently coupled triple-étalon interferometer intended for use as a focal plane instrument in the planned National Solar Observatory's Advanced Technology Solar Telescope (NSO's ATST). The algorithm converges quickly to a reasonable design that is well within the constraints imposed on the design variables, and which fulfills all resolution, signal-to-noise, throughput, and parasitic band suppression requirements.