Recent years’ advances in manufacturing technologies such as additive manufacturing (AM) have made it possible to produce engineering parts with increasingly complex shapes in a relatively short period. To address the challenges posed by such complex objects, innovative design techniques, such as generative design, have emerged in the field of engineering. In generative design, the role of the engineer is evolving from manually designing shapes to specifying product requirements that are processed by algorithms to generate a multitude of objects. Furthermore, most importantly, the physics-driven generative process is of high importance, since it assures its feasibility and suitability to fit into its operational conditions. This requires advanced numerical methods that can handle multiple interacting physical processes, including fluid mechanics, solid mechanics, and electromagnetism. Immersed boundary methods, such as the Finite Cell Method [1] (FCM), are ideal for this application due to their ability to easily handle the complexity of shapes generated by generative algorithms while minimizing the manual efforts included in simulations. In this contribution, we combine a generative design framework and the Finite Cell Method, demonstrating how the two approaches can be used together to evaluate the physical performance of generatively designed objects.