The development of the hydrogen engine as a CO2-neutral propulsion technology has gained significant attention. One challenge of hydrogen for combustion process development is the provision of a proper mixture homogenization being necessary for clean and efficient combustion. In this study, virtual product development methods are used to gain a deep understanding of the mixture formation in hydrogen engines. For example, computational fluid dynamics (CFD) studies of the CV-H2 engine show that jet guiding caps are crucial. The results show that the characteristics of the gas jet(s) must be individually adapted to the engine’s charge motion and injector installation position. To rapidly design cap geometries providing the desired characteristics of the gas jet(s), a fully automated and efficient CFD workflow of a pressure chamber is established. Design of Experiment (DoE) and Machine Learning (ML) algorithms are used to explore and exploit the high dimensional design space. The sensitivity of the geometry parameters on the jet formation can be shown in detail. The promising designs derived from the DoE are then evaluated in detail using engine calculations. Thus, the influence of different jet guiding caps on the mixture formation characteristics are evaluated. This research contributes to the advancement of CV-H2 engines by identifying the critical role of jet guiding caps in achieving optimal mixture formation. By using modern, virtual methods, the components of the H2 engine can be quickly developed to production readiness. As a next step, the use of ML algorithms will be extended to engine simulations so that jet guiding caps can be designed optimally and engine-specifically with the help of Artificial Intelligence (AI), thereby accelerating the transition to sustainable transportation.