This study applies Bayesian spatial modelling to map the HIV/AIDS epidemic in Namibia, addressing gaps in traditional surveillance methods. Using Gaussian Markov Random Fields (GMRF) and Conditional Autoregressive (CAR) models, the research estimates HIV prevalence at regional and constituency levels, through the multiscale approach accounting for spatial dependencies and data sparsity. The study utilizes 2013 Namibia Demographic and Health Survey (NDHS) data, integrating socio-economic and demographic factors. Moran’s I statistic assesses spatial clustering, revealing high-risk areas requiring targeted interventions. Findings underscore the need for data-driven HIV resource allocation, supporting policy planning and precision public health strategies to curb the epidemic effectively.