In this thesis, I investigate novel methods of sensing change in roasting coffee. Coffee roasting is a vital step in coffee’s supply chain. It is the process that takes green coffee and transforms it via a series of many chemical and physical reactions to produce the flavours for which coffee is recognised. As of today, correctly roasted coffee has been an art. My work applies different methods of sensing to track various changes in coffee as it roasts with the end goal of automating the art. The goal was to find an event or measurable phenomenon that could be used to detect when the roasting coffee has reached an ideal point, beyond which the beans would start to taste burnt. My investigations looked at various sensing methods. I expanded on three different areas of test: bean temperature, online moisture measurement using microwave resonance, and surface chemistry using Raman spectrometry. Using these different methods, I was able to produce novel and interesting measurements. Each of the measurement methods performed satisfactorily and produced novel results. These results however did not produce the desired measurable event or independent characteristics. This meant that they were not able to individually solve the problem. However, their uniqueness and application flexibility would allow them to be used as elements in future roasting sensor systems.