Buildings account for 73% of the total electricity consumption in the US. To get an in depth view of where this energy is consumed within buildings, we instrument and monitor the buildings at UCSD to study their power consumption patterns. We observe that the energy consumed is not proportional to the occupancy levels of these buildings, thus indicating energy waste. In order to make the power consumption more proportional to its actual usage, we build an occupancy detection system and deploy it in the CSE building at UCSD. Using this occupancy information as an input, we duty-cycle the different subsystems of the building to save energy. For example, we show that by careful scheduling of the HVAC system based on the occupancy levels, we can reduce their energy consumption by as much as 40%. Further, we have developed the Smart Energy Meter to monitor and actuate plug loads in the building. Our Smart Energy Meter allows us to study the energy consumption patterns on a per device basis. Based upon our smart energy meter, we have developed an analysis engine, called the Energy Auditor. It provides feedback to the users and building managers by visualizing the energy consumption data, shows them the opportunity to save energy based on the occupancy patterns and also allows the building managers to actuate the plug loads in case of a demand response event.