This work addresses the optimization of wastewater treatment plants (WWTPs) through the implementation of an intelligent agent. This agent interacts with sensors and the environment, makes decisions based on real-time data, and acts on the system to improve water treatment efficiency. Using regression models, the agent provides accurate estimates of total nitrogen in the effluent, thereby optimizing this critical process parameter. The proposed solution is adaptable and scalable, offering an innovative tool to address the complex and dynamic challenges of this type of plant environment, improving treatment accuracy and efficiency without requiring significant investment in additional infrastructure.