Abstract: This paper examines two different yet related questions related to explainable AI (XAI)practices. Machinelearning(ML)isincreasinglyimportantinfinancialser- vices,suchaspre-approval,creditunderwriting,investments,andvariousfront-end andback-endactivities.MachineLearningcanautomaticallydetectnon-linearities and interactions in training data,facilitating faster and more accurate credit deci- sions.However, machine learning models are opaqueand hard to explain, which arecriticalelementsneededforestablishingareliabletechnology.Thestudycompares various machinelearning models, including single classifiers (logistic regression, de- cision trees, LDA, QDA),heterogeneous ensembles (AdaBoost, Random Forest), and sequential neural networks.The resultsindicate that ensemble classifiers and neuralnetworksoutperform.Inaddition,twoadvancedpost-hocmodelagnosticex- plainability techniques – LIME and SHAP are utilized to assess ML-basedcredit scoring models using the open-access datasets offered by US-based P2P Lending Platform,Lending Club.For this study, we are also using machine learning algo-rithmstodevelopnewinvestmentmodelsandexploreportfoliostrategiesthatcan maximize profitabilitywhile minimizing risk.