In this thesis I discuss my work on the Arcminute Microkelvin Imager (AMI). I focus on the detection of Sunyaev-Zel’dovich (SZ) signatures at 14-18GHz. Once the background science and operation of the instrument are described I proceed to present my contribution to the calibration of AMI, including: primary beam measurements; refinements to the known antenna geometry and flagging geostationary satellite interference. This is followed by an outline of the software that I have developed to subtract sources from visibilities, concatenate data from multiple observations, simulate data, and perform jack-knife tests to evaluate the magnitude of systematic errors. The Bayesian analysis that I use to obtain parameter estimates and to quantify the significance of putative SZ detections is described. I perform realistic simulations of clusters and use these to characterise the analysis. I then, for the first time, apply the analysis to data from the AMI blind cluster survey. I identify several previously unknown SZ decrements. Finally, I conduct pointed observations towards a high luminosity subsample of eight clusters from the Local Cluster Substructure Survey (LoCuSS). For each of these I provide probability distributions of parameters such as mass, radius, and temperature. I compare my results to those in the literature and find an overall agreement.