Computing in the 21st century is rapidly moving from the personal computer model into area of ubiquitous computing, in which technology will be woven into the fabric of everyday life. This ubiquitous, pervasive computing is enabled by Wireless Sensor Networks (WSNs). Engineers deploy these WSNs in myriad applications, ranging from implanted medical monitoring devices to industrial control systems. Node energy-which translates to lifetime- and throughput are crucial metrics for evaluating wireless sensor nodes. Previously, energy reduction at the cost of performance was a common engineering trade-off for mote microcontrollers. However, increased application complexity requires greater computational power while retaining a low-power envelope. The reduced energy consumption and increased processing power enables more complex operations on collected data to be performed locally, reducing the use of energy-hungry wireless transmission systems and in turn the overall energy consumption. In this thesis, we study the use of Quasi Delay-Insensitive (QDI) circuits to design a microcontroller that fits the WSN application space. All of our architectural and circuit decisions aim to maximize the sensor node lifetime and while retaining performance in a WSN system. Security of WSN data is also critical, especially in medical or federal use cases. To address this need, we present the analysis and design of software, hardware, and hybrid AES implementations. Again, we optimized for high encryption performance while maintaining node lifetime.