Software teams often lose time to recurring or related issues because the relevant knowledge is scattered across JIRA tickets, developer discussions, and GitHub pull requests. The arXiv paper "RAG4Tickets" introduces a retrieval-augmented generation system designed to bring that information together for faster ticket resolution.

The authors propose applying RAG to the fragmented data sources that already exist in modern development workflows. By retrieving relevant context from past tickets and code discussions, the system aims to give support agents or developers the information they need without manually piecing together history.

As an arXiv preprint, the paper's full methods and evaluation are not covered in the abstract alone, so details on performance or implementation remain open. The core idea, however, is a direct response to a common operational bottleneck in software maintenance.