Rubber is one of the most versatile materials and finds numerous applications in diverse areas. The application of rubber is mostly determined by its physico-mechanical and viscoelastic properties. Rubber properties play an essential role in performing its functional requirement, which is crucial for designing a good rubber product. Therefore, the estimation and prediction of the properties of rubber and rubber composites are central to the material developers. However, many factors, such as temperature, environmental effects, and rubber formulation can influence rubber properties and make it highly non-linear. Computer simulation plays a vital role in our understanding of complex dynamics in rubber materials and provide structure-property relationship at the nanoscopic and microscopic level. An understanding of this relationship can reduce the expensive trial experiments and provide a benchmark for novel material design. Additionally, simulations at atomic and molecular levels provide the mechanism of action and the underlying physics which finally helps in designing of new materials. In the present work, all atomistic Molecular Dynamics (MD) simulation technique is utilized to predict various thermodynamic and viscoelastic properties of raw rubbers. The effect of key structural factors, that govern the properties of rubber at the molecular level, is examined using MD. In this work, we have developed the classical atomistic models for several raw rubbers and implemented methodologies for calculating their properties from MD simulations. The predicted properties using our model and methodologies are in close agreement with the experimental and available literature values. Our results establish that MD simulations are an effective tool to predict quantitatively thermodynamic and viscoelastic properties of rubber. Eventually, the same technique can be used to predict properties for crosslinked rubber, rubber composites, blends, and silica/carbon black reinforced rubbers and thus, designing a novel rubber material.