Nanoparticles filled fiber reinforced polymer composites are gaining widespread application due to the exhibition of peculiar and unique properties that can be tailored to suit specific requirements. The current paper discusses the fabrication of glass fiber reinforced epoxy-zinc oxide (EGZ) nanocomposites. The inclusion of hard ceramic ZnO nanoparticles is known to improve the mechanical, thermal, and optical properties of the GFRE composites but also tend to increase the surface roughness. Evaluating surface roughness becomes important, especially if such material is used for applications that require smooth surfaces. Thus, an effort is made to study the surface roughness of the fabricated EGZ nanocomposite using atomic force microscopy (AFM). The ZnO nanoparticles are ultrasonically blended with epoxy resin in 1, 2, and 3% weight fractions. Compression molding is used to create the nanocomposite laminates. The predictors include sonication time, compression time, and ZnO nanofiller content. The influence of nanofiller content on the amplitude functional parameter is also statistically investigated. A high-fit and prediction-ability linear regression models are designed and verified for predicting surface roughness characteristics within the experimental restrictions. All amplitude functional parameters of surface roughness were higher in nanocomposites having higher ZnO content.