The mechanical properties of cemented paste backfill (CPB) are particularly important for its application in the minerals industry. In practice, a large number of cumbersome and time-consuming experiments are required to generate the design data. To facilitate the CPB design, this study proposes an intelligent modelling framework for the mechanical properties prediction using machine learning (ML) algorithms and genetic algorithm (GA). Three advanced ML algorithms, including decision tree (DT), gradient boosting machine (GBM), and random forest (RF), were used and compared for the mechanical properties modelling while GA was used for the hyper-parameters tuning. A total of 1077 uniaxial compressive strength (UCS) tests and 231 uniaxial tensile strength (UTS) tests were performed for the dataset preparation. Mechanical properties evaluated were the UCS, the yield strength (YS), the Young's modulus (E) and the UTS. Influencing variables for these mechanical properties were chosen to be the physical and chemical characteristics of tailings, the cement-tailings ratio, the solids content, and the curing time. The results show that GA was efficient in the hyper-parameters tuning of the evaluated ML algorithms. The GBM was a good first ML algorithm for the mechanical properties modelling with high accuracy (correlation coefficients between predicted and experimental properties were 0.963, 0.887, 0.866 and 0.899 for UCS, YS, E and UTS respectively). Based on the results, a user-friendly software package, named the intelligent mining for backfill (IMB), was developed in python programming for a wider application in the minerals industry. The proposed modelling framework and the IMB will be useful for CPB design by saving time, reducing trial tests and cutting costs.