Since the optimization of analog Complementary Metal-Oxide-Semiconductor (CMOS) integrated circuits (ICs) is strongly affected by designers' experience, this paper proposes and implements an interactive genetic algorithm (iGA) computational tool with in-loop robustness analyses (corner and Monte Carlo) to perform an optimization process of operational transconductance amplifiers (OTAs) driven by human knowledge. We performed experimental studies to evaluate the optimization cycle time (OCT) and robustness of an operational transconductance amplifier (OTA), by using the iGA. Ten volunteers have participated of these studies, in which five participants were experienced designers and the remaining five were non-experienced designers of analog CMOS ICs. Our experimental results have shown that the iGA is capable of reducing the OCT in about 86% in relation to the evolution process by using a non-interactive standard GA. More interestingly, both volunteers' teams have obtained similar OCTs and OTAs' robustness, indicating that iGA might be helpful not only to avoid the common optimization issue of local minima due to the human interaction with the computational tool (iMTGSPICE), but also it is capable of producing similar results in terms of OCTs and OTA's robustness.