Resistance spot welding remains one of the most widely used joining techniques for a variety of applications. For instance, it is used in car bodies with up to several thousand weld seams. Ensuring the structural integrity of those spot welds is of crucial importance, especially for crash-relevant components. For such spot welds, a materially continuous joint is required between two or more joining partners with a specified minimum diameter. However, both in-process and post-process testing of resistance spot welds remain challenging. Either sample-based destructive inspections are necessary, or the inspection process is too time-consuming to be efficiently integrated into production. Moreover, sample-based testing only provides information on long-term deviations and is unable to detect individual weld-specific defects. Consequently, a testing method with the potential for 100 % control should be explored. This study investigates the potential of shearography as a non-destructive testing method for both in-process monitoring and structural health monitoring (SHM) of resistance spot welds. Shearography is a non-contact optical testing method based on the measurement of the deformation gradient of the surface. In the case of resistance spot welding, there is a direct correlation between the thermomechanical processes during the solidification phase after welding, the resulting deformation gradient of the surface, and the mechanical, strength-relevant properties (e.g. stiffness) of the spot weld. Two main application areas are addressed. Firstly, shearography is used for in-process quality monitoring, where geometric and mechanical joint properties are correlated with signal characteristics such as signal-to-noise ratio and gray value contrast to classify spot welds into quality categories. An AI-based evaluation system is developed to interpret the shearography data robustly, enabling real-time integration into production environments. Secondly, shearography is applied as an SHM technique, where existing spot welds are selectively excited using induction, and the structural response is analyzed to detect potential defects. The findings reveal that shearography is suitable for both in-process quality control and SHM of resistance spot welds. Analyzing the signal characteristics allows conclusions about joint quality, while AI-based evaluation increases classification accuracy (> 99.8 %) and supports transferability to other joining applications.