Many processes affecting the Earth’s weather and climate occur over space-time scales smaller than typical grid sizes of large scale models, including aerosols, clouds and precipitation, and need to be parameterized. Inadequate representation of such fast-physics processes and their interactions have been long recognized as a major source of uncertainties in model climate sensitivity and cloud feedbacks, hampering our ability to predict climate and future climate changes. Model evaluation and identification of parameterization deficiencies are essential to improving fast physics parameterization and thus advancing climate models. Over the last few decades, progress has been made in both areas of model evaluation and parameterization development, but in a frustratingly slow pace and much more and creative efforts are needed. In this seminar, the major obstacles to further progress will be discussed and approaches/strategy to overcome these obstacles to accelerate/improve model evaluation and parameterization will be explored. In particular, the multi-institution project funded by the US DOE Earth System Modeling (ESM) program, FAst -physics System Testbed and Research (FASTER), including its main objectives and key tasks will be introduced, sharing some accomplishments we have achieved so far, and challenges we still have ahead.