The optimization of the Traveling Salesman Problem (TSP) poses a significant challenge in various domains. TSP involves finding the shortest route to minimize the total distance travelled, by visiting a set of towns once and returning to the starting point, Efficient solutions are crucial for applications like route planning, logistics, and transportation. In this study, we explore an innovative approach to tackle the TSP by combining Genetic Algorithms (GA) with the K Nearest Neighbors (KNN) method. GA is employed to evolve potential solutions, while KNN helps enhance the quality of these solutions by considering the proximity of cities. Through of simulations, we fine-tune hyper parameters to develop a model capable of predicting optimal routes for our training data, with accuracy as our key performance metric. The results are based on experiments conducted using publicly available datasets from the Udemy course platform.