The reduction of anthropogenic emissions of greenhouse gases requires decreasing the overall consumption of primary energy. Thus, waste heat recovery at medium-to-high temperature is an opportunity for generating electricity while reducing the need for primary resources. Recently, supercritical carbon dioxide power cycles (S-CO2) are emerging as a promising solution. However, a method lacks to simultaneously optimize their layout and design parameters, without relying on superstructures defined a priori. To this end, this paper suggests a novel extension of the superstructure free SYNTHSEP methodology, a bottom-up approach for the optimal synthesis and design of thermodynamic cycles, to handle also super- and transcritical cycles. An Evolutionary Algorithm combining elementary cycles makes it possible to define optimal S-CO2 configurations without limiting the search space of the optimization problem. The objective consists in finding the S-CO2 topology and design parameters that maximize the mechanical power extractable from waste heat streams in the temperature range from 200 to 700 °C, typical of the industrial sector. Results demonstrate the capability of the method to find optimal cycle layouts for any given waste heat temperature, and to achieve, at the same conditions, cycle efficiencies up to 5 % higher in relative terms than the best ones in the literature.