This report presents an intelligent parking system, based on artificial intelligence (AI) and Internet of Things (IoT) that aims to solve the available parking places finding problem. It provides a real-time information about the occupation of three city parking lots. The system uses hybrid approach for monitoring parking lots with high density (more than 90%), combining computer vision (OpenCV and YOLO) for detecting vehicles and sensors. For prediction of the occupation, individually trained Bidirectional Long Short-Term Memory (BiLSTM) models have been developed. They effectively analyze timing data. The evaluation is performed under a standard data separation (80% learning, 20% validation). The achieved results show high prediction accuracy of the BiLSTM models, that makes them a reliable solution for the occupancy prediction. The mobile application offers to the users prognoses, based on timing data analysis, where will be available parking slots in the next 15 min and navigate the drivers in an optimal way, reducing the time for available parking space search.