Scour is a complex phenomenon, its complexity increases with the change in the geometry of the obstruction. Most of the investigations were carried out on the scour process at a uniform pier. However, in reality, many bridge piers behave as non-uniform depending on the exposure of their foundation into the flow field. All the experimental investigations were carried out in the present study to understand the effect of pier geometry, the position of footing top with respect to bed level, and sediment mixtures (uniform and non-uniform) on local scour under clear water condition. A total of 106 experiments were conducted in the present study with a different combination of pier models, sediment mixture, and footing top with respect to bed. A maximum scour depth model was developed using 182 data points consisting of experimental data (106) and extracted from the literature (76). To develop a model, state-of-the-art Artificial Intelligence (AI) based modeling technique known as Gene Expression Programming (GEP) was employed in this study. GEP model was developed by using 130 data points and independent 52 data points for the model validation. The performance of the proposed scour model for the compound bridge pier was found to be satisfactory.