We address the challenge of increasing malaria incidences in Sub-Saharan Africa, with a focus on Vhembe, exacerbated by climate change and limited surveillance systems. Our early warning system, MOEWS, leverages the power of machine learning while also acknowledging the coping mechanisms of local people. MOEWS is unique in its integration of indigenous knowledge, machine learning, and drought indices. We use a mobile app to collect observed IK indicators in real time and disseminate malaria outbreak alerts through both the mobile app and a web portal.