WorldPop in CARTO provides global demographic insights at high granularity, enabling various geospatial analysis applications such as human exposure to hazards and climate change impact on ecosystems. The dataset is derived from census data, fine-resolution estimates, and remotely-sensed geographical covariates. Two main methods are used: dasymetric mapping and a convolutional neural network-based approach for identifying settlements using high-resolution satellite imagery. The WorldPop dataset offers coarser resolution estimates compared to the High Resolution Settlement Layer (HRSL) but provides more detailed information on variations within settled areas. It is available globally for every year between 2000 and 2020, both at a 1 km by 1 km grid and a 100 m by 100 m grid, with additional age and gender structures data.