Understanding text automatically is key to plenty of applications such as customer feedback analysis, information extraction from documents, and content moderation. But how should we tackle these tasks nowadays? Should we just ask ChatGPT and call it a day? In this talk, I will first discuss how Large Language Models (LLMs) are changing the game for tackling NLP applications. Then, I will present and demonstrate NuMind, a modern tool that leverages LLMs to efficiently create custom NLP models (such as classifiers and entity detectors) through a paradigm inspired by how humans teach each other, which we call “Interactive AI Development”.