Traffic jams are a significant challenge for big cities, and they are one of the reasons that lead to increased travel times, fuel consumption, and environmental pollution while complicating transportation management. Traditional traffic monitoring methods often lack the precision and adaptability required for modern road systems. As a promised solution, intelligent systems powered by computer vision have emerged as a transformative solution. This study introduces a novel framework for vehicle detection, speed estimation, and traffic-aware guidance, leveraging a newly developed dataset with five vehicle classes and achieving an mAP of 96%. The framework also integrates a speed estimation with a minimal error of 2.6 km/h. Combining vehicle count and speed estimation, the framework dynamically tracks vehicle movement directions and provides optimal route recommendations to avoid congestion in real-time. By utilizing advanced computer vision with actionable insights, this approach proves its potential for revolutionizing urban traffic management and enabling more intelligent, efficient transportation systems.