Abstract
The rapid integration of large language models into software development and enterprise automation has fundamentally reshaped how organizations design, build, refactor, and operate information systems. Tools such as AI-assisted code editors and intelligent automation platforms represent not merely incremental productivity aids but structural shifts in the relationship between human expertise, software architecture, and business process execution. This study investigates the convergence of large language model–driven development environments, enterprise software engineering practices, and automated business process management. Drawing strictly on established literature related to large language model engineering, refactoring theory, Java persistence performance, enterprise systems, robotic process automation, and workflow validation, this research develops a comprehensive analytical framework that connects artificial intelligence–augmented coding tools with enterprise-scale performance, reliability, and governance considerations. The methodology follows a qualitative, theory-driven synthesis of prior empirical and conceptual studies, emphasizing architectural patterns, performance trade-offs in CRUD-intensive systems, and the evolving role of automation across supply chains and forecasting functions. The results highlight how AI-enhanced development tools influence code quality, refactoring discipline, persistence-layer efficiency, and process automation reliability. The discussion critically examines limitations, including opacity in model behavior, risks of automation drift, and the challenges of validating complex AI-driven workflows. The article concludes by positioning large language models as a foundational yet incomplete component of future enterprise systems, requiring rigorous engineering, process alignment, and performance-aware integration to realize sustainable organizational value.