The development of nanoscience and nanotechnology has a strong dependency on the advance of computational chemistry. In this context, one of the most important concepts is the basis functions set. This linear combination of functions provides an approximate solution to Schrödinger equation for many electron atoms and molecular systems. The construction of basis function is a complex task and influences on the speed and precision of the electronic structures calculus Conventional non-linear programming techniques have been extensively used in parameterization, but they cannot be used to build a set of basis functions. This work intends to propose a methodology based in Evolutionary Algorithms to parameterize and search for the best way of using Gaussian primitives in calculus of electronic structure. The advantage of using evolutionary techniques is the ability to obtain good solutions for the continuous non-linear programming problems, which are at the same time discrete. Also, there are no necessary previous knowledge of good(standard) solutions for a certain problem. This work had evaluated different ways of build basis functions with the use of evolutionary algorithms. This essay inserts an unprecedented methodology in literature to perform construction of atomic basis functions. The tool developed here was applied to build the basis functions for the following atoms: B, C, N, O, F, Ne, Na, Mg, Al, Si, P, S, Cl, Ar. All cases of the applied methodology, which uses coevolutionary algorithms, present better results than the ones described in literature.