Due to the numerous and independent parameters involved in 3D printing, additive manufacturing doesn’t guarantee reliable structures and properties. It often delivers different quality of prints and requires an expensive trial-and-error approach before finding the optimal combination of the numerous and independent input variables. Machine learning is therefore an ideal solution to this nonlinear problem and provides an informed guess on printing parameters based on a minimal set of experiments. In this dissertation, using the case example of Fused Deposition Modeling through the open source XE Pulse 3D printer, and the porosity defect, a machine learning framework is developed to predict the defect’s occurrence and the types of combinations of printing variables (Infill type and overlap, printing speed, layer height and print temperature) to ensure its minimal presence. Specimens were 3D printed using the XE pulse printer and CT-scanned using the X7000 Industrial CT X-Ray Inspection System. Raw datasets were collected in the form of grayscale image files (around 7,300 images) from the CT-scan. An image classifier was developed and trained to sort exploitable images from defective ones. Intelligent scripts were created to extract porosity features. A Multi-Layer Perceptron (MLP) was then developed and trained to predict porosity through the different layers of the 3D printed parts. Initial training and further cross-validation over smaller subsets improved the predictive capabilities of the MLP. Given the size of the dataset and input features, the model’s accuracy has proved to be optimal: The perceptron was able to predict reasonable values of porosity for established and unknown combinations of input variables for two different sets of specimens. Finally, a scalability study followed to establish the impact of scaling on the defect formation and prediction of 3D printed parts.