Chronic kidney disease (CKD) is a major global health concern that necessitates creative methods for early diagnosis and treatment. The goal of the current manuscript is to improve CKD diagnosis by creating prediction models with machine learning methods. Improving patient outcomes, enabling prompt interventions, and assisting in healthcare decision-making are the goals. Here, we collected CKD patient’s data, preprocessed it, and subjected it to feature engineering to optimize model input. Diverse machine learning algorithms, including decision trees and k-nearest neighbors, are explored and compared for their efficacy in CKD prediction. The project’s outcomes aim to deliver a robust predictive model, facilitating early detection and personalized intervention strategies. The findings are anticipated to positively impact healthcare practices, improving CKD management and patient outcomes.