In many areas in industrial engineering one may be faced with the question how an electromagnetic device has to be designed such that a both a rather complex set of requirements such as geometrical constraints has to be fulfilled, and of which the magnetic properties has to be optimal in some sense. Given an electromagnetic design, a variety of methods exist to compute the additional magnetic properties and hence verify the constraints. However, the inverse problem, in which the optimal parameters are to be calculated given a set of constraints, is in general harder to solve. In this paper we focus on quasi-static electromagnetic problems, where the inverse problem is to find a certain conductor shape confined to an arbitrary but given surface, and electromagnetic properties are prescribed. Also conductive surfaces may be present, which affect these electromagnetic properties. With some additional assumptions the inverse problem can be formulated as a quadratic optimization problem with linear constraints.