Leveraging the astounding progress in 3D printing techniques with an increase on their quality and precision, fine-tuning of metamaterials can be performed at an industrial level not only regarding its external finishing, but also on a topological level with mechanical implications. Periodic, grid-like architectures could be replaced with more irregular or intricate ones, rendering each strut unique in topological (shape, length, lay-out) and physical terms (varying mechanical properties). That way, the conditions for the appearance of certain undesired and complex mechanical phenomena (such as fatigue, creep or, especially, buckling) would not be shared by any two given members, thus localizing the issue while emulating a continuous medium despite its architectured nature. Given a node distribution and certain boundary conditions (domain size, loading case), a method for the design, univocal statistical classification and topological optimization for metamaterials is presented on this paper.