This study aimed to evaluate the potential use of elemental powders, Ni oxyreduction and Ti HDH (hydration-dehydration) to obtain the NiTi shape memory alloy in cylindrical specimens from additive manufacturing process by Selective Laser Melting with laser power and energy density under control (100 to 150 W and 25 to 40 J/mm 3 ). Scanning electron microscopy (SEM), X-ray energy-dispersive spectroscopy (EDS) and X-ray diffraction (XRD) conjugated analyses showed that all processing parameters yield samples with Ti enriched regions surrounded by alloy layers associated to Ni 3 Ti, Ni 3 Ti 2 , and NiTi 2 intermetallic. This evidence suggests that the parameters applied were not enough to promote the complete fusion of Ti particles, indicating that the samples presented a melting microstructure with evidence of defects located due to lack of fusion due to the irregular voids and Ti islands that resemble the irregular morphology of the starting Ti HDH powder, and pores depending on the retention of gases in the fusion pool. Each explored condition presented in its structure a set of different phases in nature and proportion, without the NiTi intermetallic. Also, justifies the apparent and real density values not compatible with the NiTi intermetallic theoretical density, but the density resulting from the mixture of different Ni-Ti system phases formed. Observing the decrease in the cracks and pores, and the real densities measured, compared with the theoretical density of the NiTi intermetally, the specimens that represent the best conditions are those produced with 125 and 150 W with 30 J/mm 3 , in order not compromising the SMA properties, and would allow microstructural evolution for the formation of the NiTi through heat treatment. Although the most favorable parameters, the NiTi system did not exhibited an austenitic matrix, and then the adoption of the proposed elementary powders mixture will be promising when followed by solution heat treatment, which is one of the works under development by our research group.