This paper is concerned with the description of an automatic design environment developed for the generation of optimal windmill blade profiles as well as the presentation of some results of its application. The objective is to provide a tool for the definition of wind turbine blades that are adapted to specific locations characterized by particular distributions of wind conditions. The design environment is modular and is structured into two main parts; a search module based on macroevolutionary algorithms, and an aerodynamic simulator based evaluation module developed specifically to make optimal use of computing resources while preserving accuracy. The search module works over the very high dimensional blade space to find the best individual through macroevolutionary techniques. The evaluation module is in charge of providing a fitness value for each blade through the computation of its efficiency by means of an optimized aerodynamic simulator that takes into account both the effects on wind on the blade and the effects of the other blades in the turbine. In order to produce a simulator that would lead to acceptably low computational costs, a combination of blade element theory and a panel method was used. In addition, to improve the accuracy of this approach, a processing stage that includes the application of neural networks and an integral boundary layer code was developed to correct the results obtained.