In a context of more and more stringent environmental regulations and of energy savings, ship hull efficiency has become a key design aspect for all trade boats like bulk carriers, tankers or container vessels. To achieve an efficient hull design, the resistance curve of the boat needs to be assessed as early as possible in the initial design to establish the best possible architecture in a very short time before contracting. Such an assessment remains largely made with empirical methods due to the time constraint that prevents an extensive usage of CFD solvers, but those empirical approaches remain quite limited in terms of robustness and accuracy. Also it does not provide insight on the 3D physical field like wave patterns or pressure distributions on the hull. In this paper we are presenting how the Deep Learning Physics (DLP) developed by Extrality can help address this challenge by delivering fast 3D high-fidelity predictions. First a generic DLP-model is built from a dataset of 1000 CFD simulations covering a large portion of the design space. A general description of the DLP-model training is provided. Then a validation is carried out against CFD results, at different scales, from 0D global coefficients up to 3D volume. Once validated, the DLP-model is then used to optimize the position of longitudinal center of buoyancy (LCB) of the Japan Bulk Carrier (JBC, Larsson et al. 2015), in order to achieve the minimum ship resistance. The initial hull form was modified by the Lackenby method (Lackenby, 1950). This led to a wave height in the bow portion of the new vessel significantly smaller, resulting in a reduced wave resistance. Minimum resistance was achieved for the LCB located at -0.6% relative to its baseline. While CFD calculations require tremendous time and effort without generalization and learning ability, Deep Learning Physics can predict in quasi-real time fluid behaviors to assess hydrodynamic performances of multiple designs. Such an hydrodynamic design loop takes no longer than one day to a naval architect allowing early stage optimisation.