Watch this video on YouTube
A Generative AI Engineer is tasked with developing a preprocessing pipeline for a custom LLM application that normalizes user queries and removes unnecessary symbols. Which approach should they use?