Screening of new materials from computational search allows quick characterization and development of materials from pure theory. This can allow for materials to be discovered or engineered with specific properties to great accuracy.

To reach the point of standardization of purely theoretical development and discovery of materials, increasing the throughput of atomic scale methods such as DFT and Quantum Monte Carlo becomes essential, as well as increasing the computational power available to perform such studies. Even within the realm of computation, disparities in xc-functionals affects the predicted properties of materials and reduces confidence in them. Thus, increasing the confidence in atomic-modeling methods to separate the new discoveries from errors in computation is essential to accelerate this leap.

A not-so-insignificant hurdle lies in the development of these engineered materials that may have unique phases in the lab. However, my overall opinion is that accelerated discovery of materials via atomistic simulations is a crucial part of the future of materials science.