The Pierre Auger Observatory, located near the town Malargüe in the province of Mendoza, Argentina, is the largest cosmic-ray detector in existence, covering an area of 3000 km$^{2}$. The upgraded Observatory, in Phase II of operations, consists of a surface array of 1660 stations combining water Cherenkov, scintillator, and radio detectors. A subset of stations also includes underground muon detectors. Additionally, fluorescence detectors located at four sites overlook the array. The science goals for the enhanced Observatory include the measurement of the properties of ultra-high-energy cosmic rays with large statistics and high sensitivity to the primary composition. The Observatory is also sensitive to photons and neutrinos at the highest energies, allowing it to participate in multi-messenger studies. The Auger Offline Framework provides the tools to perform detailed simulations, using the Geant 4 toolkit, of all components of the Observatory and the analysis of both data and simulated events. It proved to have the flexibility needed to evolve during the lifetime of the Observatory, to accommodate new sub-detectors and, recently, changes to the station readout electronics. A new challenge is interfacing the framework with Machine Learning tools for both the development and execution of neural-network-based algorithms. Independent of the framework, CORSIKA 7 is used to simulate particles, fluorescence light, and radio signals produced by air showers. The production of simulations is coordinated
centrally to provide standard libraries for analyses and to optimize the use of computing resources. We will describe the evolution and status of the Offline Framework and the tools used to coordinate the simulation efforts. We will also discuss the challenges of the massive simulation efforts and the resources consumed to provide the simulation libraries required by the Collaboration.