Traffic congestion significantly disrupts our daily commutes, often causing delays and preventing us from reaching our destinations promptly. One primary reason for this congestion is the number of vehicles surpassing road capacity, especially during rush hours. Previous studies have primarily concentrated on mapping urban congestion zones and forecasting future patterns. However, understanding the root causes of congestion is crucial to devising practical solutions. This research project addresses this by introducing an intelligent traffic light system powered by artificial intelligence. Leveraging the versatile Python programming language, widely recognized for its adaptability in creating AI solutions, employed convolutional neural networks (CNN) to analyze images of pedestrians and vehicles in real-time. This will enable the traffic lights to prioritize traffic flow optimally, potentially minimizing wait times and alleviating congestion. The intelligent traffic light reduced car crossing time at the intersection from 10 to 6 s and increased the throughput to 33 cars per minute compared to the conventional light’s 21 vehicles. The system’s algorithm identifies vehicular flow, optimizing traffic light changes for efficiency.