Gerber saddles represent critical structural elements in reinforced concrete bridges, as they are subjected to high stress states and degradation phenomena that may compromise their integrity and, in severe cases, lead to the collapse of the entire structure. Their vulnerability is acknowledged in the Italian Guidelines for the Classification and Risk Management of Existing Bridges. This study summarizes and presents research carried out on Gerber saddles, beginning with an analysis of their structural behaviour and extending to the assessment of various available strengthening techniques, to identify optimized strategies that consider structural, practical, and economic aspects, all within a framework of enhancing safety and infrastructure resilience. Through a critical review of the state of the art, a comprehensive database of experimental tests from the literature has been collected, which proved useful in identifying the main failure mechanisms in relation to construction details. The comparative evaluation of strengthening methods has led to the development of a logical framework for selecting the most appropriate interventions, also assessing operational impacts and on-site feasibility. Particular attention was given to the use of external post-tensioning, which has proven to be effective, cost-efficient, and minimally invasive. Through nonlinear numerical analyses and application to case studies, a design methodology was proposed that allows for the optimization of interventions based on pre-tensioning levels, significantly improving the structural performance of Gerber saddles. Moreover, the integration of Machine Learning techniques has highlighted the potential of such tools in enabling fast and reliable predictions of load-carrying capacity, providing practical support for structural safety assessments in professional engineering practice. Therefore, the research presented offers helpful tools for engineering applications, contributing to improved maintenance management and intervention planning on existing infrastructures.