Assistente virtual com inteligência artificial para acompanhamento de projetos em uma empresa de software logístico

Abstract

With the advancement of technology and the increasing complexity of project management in software companies, the use of tools capable of integrating and organizing large volumes of project information distributed across multiple platforms has become essential. In this context, this work proposes the development of a virtual assistant based on Generative Artificial Intelligence, using the RAG (Retrieval-Augmented Generation) architecture, with the goal of supporting employees and project managers in monitoring and making decisions related to corporate projects. The proposal is grounded in the application of large language models (LLMs) combined with vector databases and semantic search mechanisms, aiming to consolidate fragmented information originating from different tools used within the corporate environment. To achieve this, the assistant is designed to operate through a web interface, facilitating access to information such as schedules, tasks, resources, and performance metrics. The system’s structure encompasses both the knowledge ingestion pipeline and the response-generation flow, focusing on contextualization and usefulness for project management.