The proliferation of Internet of Things (IoT) devices and associated data streams, are en-abling a new class of dynamic data driven applications. However, processing this data in an robust, effective and timely manner can be nontrivial and presents a new set of challenges. This paper introduces Pulsar, a data-driven framework that leverages edge resources to support location and content aware processing of data streams. In this paper we present two motivating application use cases and their requirements, and then describe Pulsar, its architecture and operation. We conclude the paper with a discussion of the status of the Pulsar framework and current and planned research and development activities.