Electric vehicle (EV) adoption depends heavily on the availability and accessibility of charging infrastructure. In this work, we propose a multi-agent framework for optimizing the placement of gas and charging stations. The multi-agent system models drivers' behavior with varying goals and constraints, interacting in a shared environment. Preliminary results using a baseline genetic algorithm demonstrate the feasibility of our approach and provide insights into optimal station distributions. This framework can be extended to incorporate more sophisticated evolutionary algorithms or real-world datasets.