Accurate soil classification in geotechnical engineering remains essential for making crucial decisions during foundation design and construction. Traditional methods achieve precise results but are constrained by their extensive time commitments and labor demands. AI and ML algorithm advancements have led to new, efficient, and scalable methods for soil classification. This research applies contemporary AI and ML algorithms for soil classification according to the Indian Standard (IS) Soil Classification System through laboratory test results. The validation of this method occurred through a case study approach, which analyzed fundamental properties, including liquidity limit and grain distribution, along with soil moisture content and shearing properties of soil samples, since these properties play an essential role in soil classification accuracy. The study achieved high predictive accuracy through modern machine learning approaches, including ensemble methods, neural networks, Random Forest, and XGBoost, and also enhanced model reliability with Explainable AI (XAI).