In the present study, an ML based framework (MIGEMM) is developed to predict the local strain distribution, failure strain and the evolution of plastic anisotropy during room temperature tensile deformation of Aluminum alloy AlSi10Mg. A special emphasis has been placed on understanding the linkages between defects volume fraction, size and distribution, the observed tensile behavior, the evolution of plastic anisotropy and local failure strain measurements. For this purpose, standard ASTM-E8M subsize tensile specimens are printed using five different build directions and are subsequently scanned using 3D X-ray Nano-CT scanner to characterize the average volume fraction, shape and size distribution of defects such as porosity. Quasi-static, room temperature tensile tests are performed, and the local strain evolution, plastic anisotropy and failure strains are characterized using a digital image correlation (DIC) system. Given the laser process parameters and defects distribution, the developed AI framework shows excellent predictions for local strain evolution and r-values.