This paper introduces a system for recognizing activities in home settings using a set of small, easy-to-install, and low-cost state-change sensors. These sensors are designed to be "tape on and forget" devices, providing an alternative to more invasive methods like cameras and microphones. The system has been deployed in multiple residential environments with non-researcher occupants, and preliminary results show that it can detect activities such as toileting, bathing, and grooming with accuracies ranging from 25% to 89%. The authors aim to develop a system that can automatically detect changes in behavior patterns, which could be useful for proactive care for the elderly. The paper discusses the challenges of recognizing activities in complex home settings and highlights the potential of using simple sensors to provide powerful clues about user activities.This paper introduces a system for recognizing activities in home settings using a set of small, easy-to-install, and low-cost state-change sensors. These sensors are designed to be "tape on and forget" devices, providing an alternative to more invasive methods like cameras and microphones. The system has been deployed in multiple residential environments with non-researcher occupants, and preliminary results show that it can detect activities such as toileting, bathing, and grooming with accuracies ranging from 25% to 89%. The authors aim to develop a system that can automatically detect changes in behavior patterns, which could be useful for proactive care for the elderly. The paper discusses the challenges of recognizing activities in complex home settings and highlights the potential of using simple sensors to provide powerful clues about user activities.