Recent developments in the domain of artificial intelligence research reveal significant advancements propelled by the power of deep learning. However, the design of deep learning architectures remains a tedious job, demanding both expertise and time. To mitigate this challenge, neural architecture search (NAS) emerges as a solution, automating the process and minimizing human intervention and repetitive tasks. In this work, we employ a search strategy based on genetic algorithms (GA), a method that proves effective in unearthing innovative architectures. Notably, these architectures are structured on a layer-based foundation, showcasing a nuanced approach to optimizing the potential of deep learning systems. Through the integration of NAS and GA, our work contributes to streamlining the intricate process of designing novel and efficient AI architectures.