The exponential growth in PVT corners due to Moore’s law scaling, and the increasing complexity of mobile chips caused by an explosion in the demand for consumer applications has ushered in significant cost and power-related challenges for manufacturing mobile device chipsets within a predictable schedule. Two main reasons for this are the significant amount of human time involved in design-time decision making, e.g., related to area, power, timing, and significant increases in complexity of the human decision-making space. The problem is that human design experience is not easily replaced by design automation tools. Reinforcement Learning is an emerging AI technique that allows us to mimic human design experience and automate human decision making without loss in quality of the design, while achieving significant reduction in the design time. In this talk, we illustrate the application of Reinforcement Learning in mimicking human designers in the loop with two examples. The first example is based on chip power optimization, and the other on physical design timing closure. By minimizing the amount of human-in-the-loop decision making, not only are we able to maintain a predictable schedule despite the growing complexity, but the Machine Learning algorithms are also able to identify more optimal choices in the design space that maybe missed by a human designer.