This work aims to fully leverage the design flexibility of additive manufacturing by directly utilizing discrete simulation models, which offer the richest design space, to parameterize structural properties such as shape, thickness, and topology for gradient-based optimization. To ensure the regularity of the design in an effective and efficient manner, explicit and implicit filters are consistently applied to the structural properties. This involves introducing a control field and a map that relates it to the physical field, in order to control design properties such as surface smoothness. Consequently, the optimization problem is essentially solved in the control space, and changes are transferred to the physical space through the filtering operation. The described technique has been successfully applied to various applications across the automotive, aerospace, and civil engineering industries. We demonstrate the applicability of this procedure on structures that will be printed using materials and technologies developed at the Collaborative Research Centre TRR 277 Additive Manufacturing in Construction (AMC).