Patient-specific computational simulations of cardiovascular function critically depend on accurate, patient-specific geometric models of anatomical domains of interest. However, deriving such models from medical imaging data remains a time-consuming and largely manual process, limiting scalability, reproducibility, and clinical translation. This tutorial-style presentation introduces two complementary frameworks that address these challenges by enabling automated geometric modeling of the cardiovascular system: SeqSeg for vascular modeling [1] and MeshGrow for integrated cardiac and vascular mesh construction [2]. We first present SeqSeg, an automatic framework for patient-specific vascular segmentation and surface generation that formulates vascular modeling as a sequential local prediction problem. SeqSeg enables robust tracking of complex vessel trees from sparse initialization and has been shown to outperform global segmentation benchmark models in vessel completeness and centerline coverage [1]. Recent extensions include fine-resolution modeling, support for multiple vascular anatomies, minimum vessel radius and subvolume constraints, initialization from pre-existing centerlines, and automatic vessel capping for boundary condition specification. Practical advances supporting adoption include a 3D Slicer plugin, released pretrained weights, and a Python package for integration into simulation workflows. We then introduce MeshGrow, an automated framework for unified cardiac and vascular mesh generation [2]. MeshGrow integrates learning-based cardiac chamber modeling via LinFlo-Net, enforces non-intersecting mesh elements during mesh prediction, and enables automatic construction of physiologically meaningful interfaces, including valve boundaries such as the left ventricle--aorta junction. Quantitative evaluations demonstrate that MeshGrow produces simulation-ready meshes with accuracy comparable to or exceeding state-of-the-art methods while substantially reducing manual modeling effort [2]. We conclude that the combined SeqSeg--MeshGrow pipeline enables scalable, reproducible, and fully automated generation of cardiovascular geometries suitable for large-cohort studies and patient-specific simulations. [1] Sveinsson Cepero, N., Shadden, S. C. Annals of Biomedical Engineering, 2025. [2] Sveinsson Cepero, N. et al. FIMH 2025, 2025.