The field of cybersecurity is constantly evolving, and Device Fingerprinting (DFP) has emerged as a crucial technique for identifying network devices based on their unique traffic data.This is necessary to protect against sophisticated cyber-attacks. However,automating device classification is complex, as it involves a vast and diverse feature space derived from various network layers, such as application,transport, and physical. With the advances in machine learning and deep learning, DFP has become more accurate and adaptable, integrating multi-layered data and emphasizing the need to balance robust security measures. The study of DFP, especially in the context of emerging protocols like HTTP/2 and HTTP/3,remains a critical area of research in cybersecurity. This talk focuses on enhancing real-time threat detection while navigating the challenges of scalability.