Pervasive computing promotes the integration of smart electronic devices in our living and working spaces to provide advanced services. Recently, two major evolutions are changing the way pervasive applications are developed. The first deals with moving computation and storage to the edge. The second is the massive use of machine learning techniques to build these applications. However, architectural principles and integrated frameworks are still missing today to successfully and repetitively support application developers in the creation of edge-level AI applications. In this paper, we present a novel architecture and platform allowing the development of such applications in smart spaces.