Climate change and the resulting national as well as international political regulations require companies to systematically reduce their products; carbon footprint. The design of parts using multi-material design offers the possibility of optimally leveraging each materials properties, thus enabling this challenge to be met in a targeted manner. However, multi-material design has additional difficulties, as different materials have to be joined. One way to address this is through form-fit methods, which have great potential because they are mostly independent of the choice of material. One arising form-fit process uses pin interfaces manufactured by Cold metal transfer welding (CMT). This work outlines an optimization framework to support product developers in the design of such pin interfaces. Global optimization methods, such as evolutionary algorithms, are used for this purpose. These require many computations of the system response by means of FE methods. However, since the pins and the part have different length scales, leading to a very high number of degrees of freedom in the FE, a homogenization approach for such pin interfaces is currently being developed. The proposed optimization approach, including the homogenization of the pin interface, consists of the following steps: (1) Determination of the dilute strain concentration tensor and other properties for a single pin in a sufficiently sized FE unit cell, (2) computation of the elastic constants of the surrogate model for a selected pin configuration (geometry and density) using a Mori-Tanaka model, (3) determination of the system response during optimization applying the surrogate model, and (4) validation of the final part design employing a complete FE model. Our current research investigates the accuracy of the homogenization approach in comparison to full-scale FE models. We hypothesize that the optimization routine will provide practical insides into the mechanical part design, even at slightly decreased overall accuracy due to the surrogate interface material. Therefore, the design process of complex multi-material components is accelerated and can even be partially automated.