With the advancement of computer science and machine learning algorithms, machine learning techniques and solutions have become viable in detecting diseases with provided data at acceptable rates of errors. A doctor can detect disease at a much faster rate utilizing machine learning techniques. The dataset for CKD detection was taken from UCI repository. Many classification algorithms were tested including SVM, DT, KNN, and RF. These models have been thoroughly tested implying different hyper-parameters to ease the learning by the models. The null values in the numerical part of the dataset were filled with random value imputation. The nominal part of the dataset was filled with mode Imputation. Along with multiple machine learning algorithms, only the Random Forest algorithm was able to achieve accuracy of 99% average with maximum accuracy of 100% with K-fold cross-validation. With the models built, the client–server system architecture was built around it for general use purposes.