Keywords
what, is, genai, good, for

Technologies
googlescholar, microsoftforms, jupyternotebook, excel

This project investigates the effectiveness of Generative Artificial Intelligence (GenAI) in supporting software engineering tasks across different phases of the Software Development Life Cycle. It examines whether GenAI performs differently for structured tasks with clear solutions, such as debugging or test generation, compared to judgment-based tasks requiring reasoning, prioritisation and trade-off analysis.

A quantitative experimental design is used, where participants complete software engineering activities with ChatGPT assistance. Tasks range from coding activities to higher-level design evaluation and decision-making. Performance is measured using objective criteria such as correctness, along with human-centred measures including confidence, trust and perceived difficulty.

The study aims to identify where GenAI provides reliable support and where human expertise remains essential, contributing to understanding how GenAI can be responsibly integrated into software engineering practice.