The paper presents the use of an Artificial Intelligence technique dedicated to find the near optimal design of a robot manipulator which transfers silicon wafers for the electronic manufacturing. This task requires the manipulator to move as smooth as possible. The environment where the manipulator works must be free from airborne particles that are able to damage the production of very integrated electronic circuits. Poor mechanical interactions, resulting from rough and jerky motions, ease the production of these destructive micro particles. There are some strategies providing the very necessary smooth motions: by means of trajectory planning or by means of mechanical design. The paper focuses in finding the near optimal design taking in account the link length ratio. Due to the huge search space, it is necessary the use of an intelligent, robust and suitable method. A Genetic Based Search Technique has proved a suitable performance in finding the best manipulator for the mentioned task.