Force fields have proven to be essential tools for studying properties of materials at different length and time scales. Classical force fields describe interatomic interactions using classical equations, enable accurate simulations at a lower computational cost. Recently, driven by higher data availability and powerful high performance computers, a new class of precise and efficient interatomic potentials, such as machine learning interatomic potentials (MLIPs), emerged. This approach marks a paradigm shift, replacing physics-based functional forms of potentials with highly complex mathematical forms based on neural networks. These networks, trained on reference ab-initio datasets, can reproduce high-level quantum mechanical calculations with negligible errors, achieving accuracies comparable to experimental uncertainties. This work showcases the implementation of advanced artificial intelligence methodologies across a range of materials and processes relevant to IMM research.