LANGUAGE MODELS FOR AUTOMATING DOCUMENTATION AND SUPPORTING KNOWLEDGE MANAGEMENT IN REMOTE TEAMS IN IT-COMPANIES

Abstract

The article examines the potential of Large Language Models (LLMs) for automating technical documentation and supporting organizational knowledge management processes in IT projects, particularly those implemented within remote and distributed work environments. The growing complexity of modern software systems and the inherent challenges of collaboration across different time zones necessitate new approaches to standardize documentation practices and enhance internal communication efficiency. The relevance of this study lies in addressing the critical need for reliable mechanisms to capture, update, and retrieve expertise within decentralized teams, thus mitigating knowledge silos and reducing cognitive load on developers. The research employed a structured methodology, combining the method of systematic literature analysis to comprehensively review contemporary AI integration approaches in documentation workflows, the typological method to categorize and define distinct models of AI-assisted software engineering, and the modeling method to propose an implementable framework based on the Retrieval-Augmented Generation (RAG) architecture. The core findings highlight how AI technologies optimize the Software Development Life Cycle (SDLC), specifically by automating initial drafts, ensuring consistency between code and documentation, and assisting in the coordination among remote team members. Special attention is given to the RAG architecture as an efficient way to build internal, context-aware knowledge systems, allowing LLMs to access and synthesize information from an organization’s proprietary codebase and internal documentation. The proposed implementation model strategically positions LLMs not as a definitive replacement for human expertise but as a cognitive augmentation tool that handles repetitive, data-intensive tasks. This approach enables human technical writers and engineers to focus on strategic validation, contextual accuracy, and quality assurance of the generated content. Ultimately, the use of AI-driven language models is demonstrated to accelerate knowledge transfer, improve documentation quality and consistency, and enhance overall project transparency and collaborative productivity in the demanding context of remote IT management.