In modern agriculture, efficient crop selection is crucial to optimize yield and address growing food demand. Traditional methods lack precision and often do not adapt to changing environmental dynamics. Thus, the analysis of a zone’s agroclimatic conditions contributes significantly to deciding the right crop for the right land in the right season to obtain a better yield. In this work, we present an intelligent crop recommendation system that uses machine learning to recommend the best-suited crops based on weather forecasts and soil characteristics. This system can be used for informed decision making in the smart agriculture ecosystem. Our approach integrates two models: a Long-Short-Term Memory (LSTM)-based weather prediction model that forecasts future conditions such as temperature, humidity, rainfall, and other climatic factors, and a random forest classifier that uses these predicted weather parameters along with soil data (pH, nitrogen, phosphorus, potassium) to determine the optimal crop. The system provides dynamic and data-driven insights that help farmers make informed decisions about crop selection. We also performed the SHapley Additive Explanations(SHAP) analysis to explain the influence of various environmental and soil factors on crop predictions, maintaining interpretability. This study highlights the prospective of machine learning to advance precision agriculture and promote sustainable crop production in response to changing climatic conditions.