The industry for 3D printing is increasing rapidly due to its numerous advantages, including enhanced safety, cost-effectiveness, faster construction, reduced waste, complex geometries, and environmental friendliness. Mechanical qualities such as flexural strength and tensile strength are crucial when printing parts with variable process parameters for diverse applications. To optimize manufacturing processes, predictive models are essential; otherwise, the task can be costly. The tensile behavior of carbon fiber-reinforced polylactic acid specimens made by fused filament manufacturing is predicted using categorical boost, extreme gradient boosting, and decision tree regression. To determine their microstructural integrity, the specimens were subjected to stringent tensile testing and fractography examination. The resulting dataset was then utilized to gauge how well the machine learning regression methods performed. Layer height (0.1 mm, 0.15 mm, 0.2 mm), feed rate (20 mm/s, 40 mm/s, 60 mm/s), and raster angle (0°, 45°, 90°) were among the 27 possible combinations of process parameters that were examined. The results directed that the categorical boost regression was the most effective in predicting tensile strength, achieving coefficients of determination value 0.99. SHapely Additive ExPlanation (SHAP) analysis further highlighted the importance of each feature in tensile strength prediction.