Advanced nuclear power systems, spanning from optimised Boiling Water Reactors (BWRs) to Generation IV (Gen-IV) reactors play a crucial role in the future carbon-neutral energy mix. The development of these new systems require a thorough investigation of the material properties under extreme reactor conditions. For this purpose, this project utilises peta-scale computational resources to investigate fundamental degradation and protection mechanisms of key nuclear materials, from advanced fuels and structural materials to protective cladding coatings. The research is divided in two tracks, addressing both near-term safety and long-term sustainability goals. The first track focuses on the development of Accident Tolerant Fuels (ATFs) for current BWRs systems. Utilising state-of-the-art Density Functional Theory (DFT), we investigate transitional-metal nitride coatings (such as CrN, NbN, CrNbN) applied to zirconium alloy cladding. By rigorously modelling surface adsorption, oxide formation, and thermodynamic stability at the atomic level, we aim to investigate the fundamental mechanisms of oxidation resistance. The second track addresses materials development for Gen-IV and fusion applications. We investigate radiation driven processes in oxide and nitride fuels which are essential for reactor systems designed to burn reprocessed spent fuel, thereby increasing efficiency and significantly reducing radioactive waste lifetimes. Utilising first-principles methods (such as DMFT and DFT+U) to capture strong electron correlations, we characterise these fuels for future licensing. To bridge the gap between atomic-level physics and macroscopic material behaviour under varying levels of neutron irradiation, we employ a multi-scale modelling framework. First-principles physics are coupled with stochastic long term evolution models and machine learning (ML) methods to build robust, generic models for microstructural evolution. Executing these electronic structure calculations, surface chemistry models, and coupled ML frameworks in only possible with peta-scale computing. Even with a medium scale NAISS allocation we cannot complete these studies but with combination of allocation at other HPC resources ensures a reasonable workflow.