Agriculture is considered among main pillar of Moroccan economy contributing around 11% to the GDP. However, it faces growing challenges, mainly those related to water scarcity, energy dependence, and inefficient irrigation practices. Traditional systems often lack real-time adaptability and energy autonomy, leading to unsustainable freshwater consumption. This paper introduces MGFARM, an edge AI-driven irrigation platform’s framework integrating IoT, Bigdata technologies and AI-driven decision-making, for the development of smart applications. For instance, dynamically optimizing irrigation based on real-time environmental data, ensuring efficient water distribution while operating as much as possible on renewable energy sources (e.g., solar, wind). Experiments have been conducted and preliminary results demonstrate the efficiency of the proposed framework in improving agricultural practices.