Oligosaccharides are a group of complex glycans that are present in the milk of most mammals. However, human milk is unique as the concentrations of human milk oligosaccharides (HMOs) are much higher than those of other mammals, and their structural composition is more complex and varies between women. These observations prompt several questions: (i) Why are humans unique when it comes to milk oligosaccharides? (ii) Which maternal genetic and environmental factors drive the interindividual variation in HMO composition? (iii) What are the short- and long-term health benefits for the infant - and potentially also the mother? The combination of genome-wide association studies, milk transcriptomics, in vitro gene editing, and in silico pathway modeling allows us to reconstruct HMO biosynthetic pathways. Using new data mining approaches and leveraging samples and metadata from large mother-infant cohorts enable us to identify associations between HMO composition and infant and maternal health outcomes. Suitable preclinical models and clinical intervention studies allow us to corroborate the established associations for causal relationships and test for in vivo efficacy in humans. Knowledge generated from these different approaches will help us establish true structure-function relationships and provide the rigorous evidence required to improve infant health and development.