Arquitetura Ai2learn de Ia Generativa Para A Criação de Conteúdo Educacional Com Gcp

With the advent of Large-Scale Language Models (LSML), contemporary education has undergone a huge transformation. The ease of access to this type of system has allowed educators and students to use this technology in their educational activities and become deeply involved with it. Due to these paradigm shifts and the difficulties that exist in the current teaching model, such as personalization and individual student monitoring, the question of the place of these tools in the classroom has increasingly taken over the narrative, simultaneously gaining more ground in adoption on both sides. One of the most pertinent questions in this scenario turns out to be how to implement a system that maximizes the generative and customizable capabilities of LLMs in a way that fills gaps and brings more value. This paper presents a generative technological architecture focused on assisting pedagogy, using artificial intelligence tools from the Google Cloud Platform (GCP), which make it possible to create different types of educational resources based on existing content using techniques such as augmented generation by search (RAG), allowing for parameterized personalization of the creation process and content, and machine-student interaction based on educational resources.