This chapter is devoted to solving the problems related to the automation of the IC parametric optimization process and developing new algorithms. One of the most important and time-consuming phases of IC design is parametric optimization. It boils down to a selection of physical dimensions of individual components, so that necessary measurements of output signals of the circuit satisfy the imposed constraints. The complexity of the process is conditioned by the increase of the influence of secondary factors on the normal operation of circuit components, among which are parasitic elements and leakage currents. That circumstance, coupled with the need for similar ICs to be embedded into different environments and to satisfy behavioral constraints, creates a situation where similar circuits need to be optimized for different processes and working conditions. Existing precise methods are based on the transformation of the given design constraints into an optimization problem and numerous IC simulations. However, these approaches are still not sufficient from the standpoint of practical applications. A need arises to develop new methods for automated parametric optimization of ICs, which, by applying new search approaches and performing a small number of simulations, will discover an IC candidate satisfying the proposed behavioral constraints. Means of automated parametric optimization of ICs have been proposed, which transform the given circuit and constraints into a single-objective or multi-objective optimization problem, apply swarm intelligence (SI) optimization algorithms (OAs) and machine learning strategies with the goal of finding the global minimum with the least number of simulations.