NearMe is a server, algorithms, and application programming interfaces (APIs) for clients equipped with 802.11 wireless networking (Wi-Fi) to compute lists of people and things that are physically nearby. NearMe compares clients’ lists of Wi-Fi access points and signal strengths to compute the proximity of devices to one another. Traditional location sensing systems compute and compare absolute locations, which requires extensive a priori calibration and configuration. Because we base NearMe entirely on proximity information, NearMe works “out of the box” with no calibration and minimal setup. Many “location-aware” applications only require proximity information, and not absolute location: examples include discovering nearby resources, sending an email to other persons who are nearby, or detecting synchronous user operations between mobile devices. As more people use the system, NearMe grows in both the number of places that can be found (e.g. printers and conference rooms) and in the physical range over which other people and places can be found. This paper describes our algorithms and infrastructure for proximity sensing, as well as some of the clients we have implemented for various applications.