The development of the human embryo is arguably the most complex process that we could care to study. In this process, the developing embryo must undergo proliferation, reorganization, lineage diversification, and dozens of cell fate specification events. During this time, a myriad of events are happening in parallel at the cell level, each one setting the foundation for the emergence of increasingly complex tissues of increasingly complex function. Understanding the mechanisms guiding these processes is pivotal not only for embryogenesis-related applications in fertility and development, but also for regenerative medicine applications such as the development of organ replacements. In this dissertation, I propose an integrative approach to the study of morphogenesis and patterning, specifically in the context of stem cell-based models of human development. Firstly, I present a novel machine learning-assisted imaging pipeline that permits the careful characterization of cell-level events occurring in our in vitro model of epiblast cyst morphogenesis. Secondly, I present a novel agent-based model (ABM)-genetic algorithm (GA) framework for the generation of models of morphogenesis. The framework was first tested to determine its ability to generate structures of desired patterns. It was then applied for the generation of models that plausibly capture mechanisms at work during epiblast cyst morphogenesis and symmetry breaking. With preliminary in silico experiments, I showed that the framework was able to output models that partially captured the effect of initial cell number on final cyst composition. I further showed that correct structure formation was heavily impacted by just a few model parameters. Combined with in vitro experimentation, these tools have the potential to shed light into the mechanisms guiding growth, movement, and cell fate specification in in vitro models of human development.