IT project infrastructure setup automation with help of large language models

This study conducts an analysis of existing large language models (LLMs) and AI agents, identifying Llama 2 as the most suitable model for automating IT project environment configuration. A mathematical model of the proposed method is introduced to automate IT infrastructure setup and reduce development time. The system architecture incorporates modules for natural language processing (NLP), configuration generation and command execution. The effectiveness of the method is evaluated through experiments across five key production scenarios, comparing two approaches: traditional infrastructure configuration tools and the proposed LLM-based method utilizing Llama 2. Experimental results demonstrate that the proposed method reduces configuration time up to 60%, decreases error rates from 25% to 8% and improves configuration quality approximately in 3 times. The article is relevant to IT professionals engaged in automating development and infrastructure configuration processes, as well as researchers exploring the application of artificial intelligence, particularly large language models, in the IT industry.