Although wireless sensor network applications are still at early stages of development in the industry, it is obvious that it will pervasively come true and billions of embedded microcomputers will become online for the purpose of remote sensing, actuation and sharing information. According to the estimations, there will be 50 billion connected sensors or things by the year 2020. As we are developing first to market wireless sensor-actuator network devices, we have chance to identify design parameters, define technical infrastructure and make an effort to meet scalable system requirements. In this manner, required research and development activities must involve several research directions such as massive scaling, creating information and big data, robustness, security, privacy and human-in-the-loop. In this study, wireless sensor networks and Internet of things concepts are not only investigated theoretically but also the proposed system is designed and implemented end-to-end. Low rate wireless personal area network sensor nodes with random network coding capability are used for remote sensing and actuation. Low throughput embedded IP gateway node is developed utilizing both random network coding at low rate wireless personal area network side and low overhead websocket protocol for cloud communications side. Service-oriented design pattern is proposed for wireless sensor network cloud data analytics.