Internet of Things (IoT) in conjunction with Vehicular Ad-hoc Networks (VANETs) is introducing new opportunities ofintelligent and safer transport. Although the majority of Advanced Driver Assistance Systems (ADAS) are aimed at addressing the issue ofexternal hazards, they do not pay much attention to the internal condition of the driver, including fatigue, alcohol impairment, orunauthorized access. In this paper, I would describe a system that combines board-mounted multi-sensors monitoring, such as biometricaccess control, alcohol detection, and vision-based drowsiness analysis with the VANET/V2X communication. The intent is to reach forsafety cooperative interventions in a timely manner beyond isolated in-vehicle responses. We introduce the architecture of the system inlayers, flow of information and decision policies with special focus to the important issues like latency, privacy and reliability. Theframework by adopting new requirements of V2V and V2I communication sets the foundation of scalable, real-time safety infrastructuresthat can be tested on embedded systems and can be expanded to C-V2X/5G.Keywords: VANET, V2X, IoT, Multi-sensor fusion, Driver monitoring, Edge AI, Embedded systems, Vehicular safety