<p>As the transition toward electrified transportation gains momentum, the challenge of managing electric vehicle (EV) charging in real-time—while minimizing energy costs and maintaining grid stability—has become increasingly critical. Existing centralized control methods often struggle to adapt to the rapid influx of heterogeneous data, including traffic conditions, dynamic grid status, and user-specific preferences. To address this, we introduce an Edge-AI based optimization framework that manages EV charging by simultaneously considering multiple criteria in real-time. Our system synthesizes live data from traffic networks, electric utilities, and user mobility patterns to deliver context awareness charging recommendations. The framework employs a hybrid algorithm combining Reinforcement Learning with the Analytic Hierarchy Process (AHP) to evaluate and prioritize conflicting objectives such as reduced wait time, minimized grid strain, lower electricity costs, and decreased emissions. By shifting decision-making closer to the source through edge computing, the model achieves low-latency responsiveness and enhanced resilience against connectivity failures. We tested the framework using a combination of real-world and synthetic datasets to simulate diverse urban conditions. Results indicate up to 23.5% shorter wait times, 18.2% improvement in grid utilization, and 21.7% cost reduction compared to conventional approaches. These findings suggest that edge-intelligent, multi-objective systems can play a pivotal role in developing adaptive, scalable EV charging infrastructures, especially in alignment with U.S. smart grid and sustainability initiatives.</p>