This paper outlines approaches and prospects for the development of genetic and management technologies to address short and long-term effects of climate change on the livelihoods of cassava farmers. Uncertainty has always been a way of life for farmers. Further, they are one of the most vulnerable groups likely to be affected by the looming uncertainties of climate change. The long-term climate trends - drier or wetter, warmer or cooler - are what get the attention in the press, and the most consideration at the policy level. But for farmers, their compelling anxiety is more often about how climate change can increase short-term uncertainty - within a growing season and from one year to the next. Current climate change models have limited power to address these seasonal variations. There are no models of any reasonable reliability that tell a farmer what to expect throughout a growing season, especially when that growing season is for nine months to a year or more, such as for cassava. These short term challenges need to be a high priority for research, while also taking into account projected long-term trends that include higher temperatures, change of long-term precipitation patterns and higher levels of atmospheric CO². Among the most effective of these responses will be biologically-based, effective and low-cost solutions such as genetically improved varieties and integrated pest management. While breeders have a long experience in raising yield potential and selecting for tolerance to predictable drought, the new challenge will be to combine yield potential with resilience in the face of unpredictable variations in the production environment. A resilience strategy has two broad fronts. One is to incorporate specific resilience traits, and the other is to provide a wide range of variability to farmers, in order to cover an array of biotic and abiotic stress scenarios. Climate change means higher research investment if new technologies are to adequately address