This project aims to assess the mechanical quality of additive manufactured (AM) metal parts by vibrational testing. The first phase of the project involves the design of a flexible experimental methodology to capture linear and nonlinear vibrational phenomena in a fast manner (less than 1 second). Simultaneously, a numerical model will be setup to get in-depth insight into the (nonlinear) vibrational dynamics. Building on this, the project will interpret vibrational phenomena and establish a clear link between material quality indicators and observed vibrational behavior. The final phase focuses on the use of unsupervised machine learning techniques in order to distinguish between high- and low-quality AM metal parts. The unique character of this proposal is its integration of experiment, simulation, and machine learning, allowing us to get insight into the mechanical quality of AM metal parts by using vibrational testing. It is expected that this project will leverage the wider use of AM metal parts in high-tech industry.