Higher data rates and reduced power consumption have become more and more necessary in recent years due to the quick development of wireless communications. The increase in the hardware complexity to implement the new performance standards poses several challenges. Moreover, mathematical models will no longer be feasible since they cannot accurately describe communication systems due to a high degree of unpredictability. AI/ML techniques have become increasingly popular as they provide elegant solutions to the problems stated above. These techniques can be used to optimize hardware use, as well as to accurately model communication systems. This paper discusses the applications of deep neural networks (DNNs) in channel estimation and pilot signal design.