The growing demand for sustainable heavy-duty transportation has accelerated the development of hydrogen-powered Fuel Cell Electric Vehicles (FCEVs). Effective energy management is critical to maximize the driving range, preserve fuel cell durability, and ensure stable battery operation. Traditional rule-based and Model Predictive Control (MPC) strategies, while well-established, often fail to adapt to the highly dynamic power demands of real-world driving cycles, leading to suboptimal hydrogen consumption and increased fuel cell degradation. This thesis presents the design, implementation, and validation of a Soft Actor-Critic (SAC) reinforcement learning controller for the energy management system (EMS) of a heavy-duty FCEV based on the Iveco Daily H2 platform.