Additive Manufacturing, in great part due to its huge advantages in terms of design flexibility and parts customization, can be of major importance in maintenance engineering and it is considered one of the key enablers of Industry 4.0. Nonetheless, major improvements are needed towards having additive manufacturing solutions achieve the quality and repeatability standards required by mass production. In-situ monitoring systems can be extremely beneficial in this regard, as they allow to detect faulty parts at a very early stage and reduce the need for post-process analysis. After providing an overview of Additive Manufacturing and of the state of the art and challenges of in-situ defects monitoring, this chapter describes an in-house developed system for detecting powder bed defects. For that purpose, a low-cost camera has been mounted off-axis on top of the machine under consideration. Moreover, a set of fully automated algorithms for computer vision and machine learning enables allow the timely detection of a number of powder bed defects along with the layer-by-layer monitoring of the part’s profile.