In the context of materials design and high-throughput computational searches for new thermoelectric materials, the need to compute electron and phonon transport properties renders direct assessment of the thermoelectric figure of merit (zT) for large numbers of compounds challenging. On the other hand, recent discoveries demonstrate how entirely new material systems can lead to the disruption of existing technologies and a three-fold increase in the efficiency of thermoelectric generators. This chapter reviews recent efforts in developing robust computational approaches to screen the chemical space of inorganic materials for thermoelectric performance. The starting point is Boltzmann transport theory and the description of various strategies used in high-throughput computations to overcome the limitations associated with direct calculations of electron and phonon relaxation times. Next, we describe the implementation of these strategies and the resulting identification of new promising thermoelectric materials. Particular emphasis is on experimental validation of computational predictions. Finally, we discuss the current outstanding challenges including dopability of semiconductors, finite temperature phenomena, and systems beyond Boltzmann transport theory. Addressing these will further improve the reliability of predictions and bring us closer to the true thermoelectric materials by design.