Deep tight sandstone reservoirs are the major targets of hydrocarbon exploration in the eastern South China Sea,but it is challenging to predict deep sweet spots using routine methods.To evaluate potential deliverability of low-permeability reservoirs and discover those reservoirs with economic deliverability,it is necessary to identify sweet spots from the perspectives of geology and engineering.In the case study of Paleogene tight reservoirs in Lufeng area,the eastern South China Sea,we use the following methods for geologic and engineering sweet spotting.①Seismic inversion based on machine learning is employed to establish sand-stone distribution,or geologic sweet spots,in dominant facies belts;②the brittleness index model is built for seismic prediction u-sing Poisson's ratio and Young's modulus derived from rock mechanical experiments;③ prestack elastic impedance inversion is performed to obtain Poisson's ratio and Young's modulus for engineering sweet spotting.Using these methods,we locate tight sandstone sweet spots in Paleogene reservoirs for drilling site deployment.Oil production obtained after offshore fracturing demon-strates the feasibility of these methods in Lufeng area.