Chemical space captures the whole spectrum of possible small molecules and their physicochemical, biological and structural characteristics. Latest advancements in computational chemistry, generative artificial intelligence and cheminformatics have transformed ways in which vast molecular universe is visualised, prioritised and navigated for drug discovery. Molecular descriptors act as a quantitative base for these studies that captures features vital for property prediction, molecular similarity and activity modeling assessment. The incorporation of cheminformatics pipelines, machine learning frameworks and structure-based docking simulations has made rapid identification of biologically relevant subspaces possible within immense compound libraries. Generative AI methods, encompassing latent-space optimisation, reinforcement learning, diffusion models, ADMET profiles and selectivity. Databases put together from fragment-based libraries and natural products remain vital for mapping chemical diversity and improving novelty of scaffold. Together, these AI-driven and computational techniques connect data-rich chemical landscape with mechanistic and predictive modeling, speeding up the discovery of therapeutically synthetic and viably accessible molecular entities.