Conformal prediction is a technique to quantify uncertainties for an AI system by basing an input on the estimation of a prediction interval in regression problems and a set of classes in classification problems. This paper examines the application of conformal prediction techniques in predicting intervals using distinct datasets for housing prices and birth weights. By integrating quantile regression forest and LightGBM models, we aim to provide statistically significant prediction intervals. These intervals offer valuable insights for risk assessment and decision-making processes in real estate and healthcare sectors, respectively. We apply conformal prediction intervals with Winkler interval score that covers the true values with high probability.