Software Development using Low-Code Platforms or AI-Assistants: A Case Study-based Comparison
TL;DR
The results suggest that LCPs with integrated AI enhance development efficiency and predictability compared to traditional LCPs and purely AI-assisted development.
Abstract
Low-Code Platforms (LCPs) and AI assistants accelerate and democratise application development by generating artefacts through visual tools, drag-and-drop components, prebuilt logic, and large language models (LLMs) for professional and citizen developers. While prior work has examined these approaches in isolation, the comparative use of these approaches for developing applications remains insufficiently understood. This paper presents a case study, in which multiple teams developed a web application based on the same requirements over 13 weeks using LCPs, LCPs with AI integration, or AI-assisted development. The study combines report and repository analysis with longitudinal survey data to capture technology usage, development processes, and developer perceptions. This primarily qualitative study aims to identify new insights into the potential and limitations of web development with LCPs and AI assistants. Our results suggest that LCPs with integrated AI enhance development efficiency and predictability compared to traditional LCPs and purely AI-assisted development.