Neurodevelopmental disorders (NDDs) begin in the earliest months of life, but the clinical manifestations of many such disorders—including intellectual disability, autism spectrum disorder, attention-deficit hyperactivity disorder, and fetal alcohol spectrum disorder—are often not seen until much later. Theoretically, very early intervention provides the best chance for optimizing outcomes for children at risk. However, opportunities for early intervention may be thwarted by a dependence on assessing overt phenotypic or behavioral symptoms that often emerge later in childhood. In this article, we examine the utility of a transdiagnostic, dimensional approach to very early identification and intervention for infants at risk of NDDs. We consider how insights from epidemiology, systems biology, and developmental neuroscience may be integrated to aid identifying at-risk infants and developing targeted, adaptive interventions. We also highlight directions for research.