The emergence of applications operating independently of direct human control is inevitable. Research on high-level models for this class of applications— e.g. on autonomous agents and distributed AI— has revealed the shortcomings of current architectures and middleware interaction paradigms. If we focus on complex real-time systems made of embedded components, that evidence is even more striking. Event models have shown to be quite promising in this arena, but they often lack one or several of the following key points: seamless integration with a programming model; adequate layer structure; and the provision of support for non-functional attributes, such as timeliness or quality of service, or security. This paper discusses infrastructural support to construct large-scale proactive applications based on the use of real-time sentient objects, and is specially devoted to the latter two points.