In times when companies must respond efficiently to market demands, reverse engineering plays a fundamental role. Although design processes are commonly developed with digital workflows, in reverse engineering different and independent phases with bottlenecks between them are involved. The present work addresses the challenge of establishing a continuous and efficient data flow between the three-dimensional digitized data obtained with 3D scanning and the automatic generation of NC toolpaths. A methodology is developed for the generation and optimization of NC toolpaths directly from the 3D point cloud data obtained through the three-dimensional scanning of pre-existing geometries. The methodology consists of an algorithm developed with Grasshopper, a script for visual programming in Rhino’s interface. It does not only attempt to reconstruct the three-dimensional geometry of the scanned part but also, it directly generates the tool paths and optimizes them with evolutionary optimization algorithms that are integrated in the methodology. A case study is developed for TMU-SIO TOWGTAI machining center with the proposed methodology. Finally, the obtained results and the efficiency of the methodology are analyzed and presented.