This paper introduces SmartSimVR, a ground-breaking research initiative focused on the development of a user-specific intelligent architecture for immersive virtual environments. The primary objective of this architecture is to address real-time artificial intelligence training and adapt the virtual environment based on the user's state or external parameters. In a case study centered around the detection of cybersickness, an undesirable side effect in immersive virtual environments, we employed this architecture in a driving simulator application. Leveraging the capabilities of this architecture enables the optimization of virtual reality experiences for individual users, resulting in increased comfort.