New Research Agenda for Prompt Engineering in Software Development

A new paper submitted to arXiv on 2 September 2026 has emerged from the First International Workshop on Empirical Prompt Engineering for Software Engineering (PROMPT-SE), co-located with EASE 2026. The research, produced through structured community discussion, outlines a comprehensive research agenda for the field.

Current Applications Across Software Engineering

Prompt engineering is increasingly used across Software Engineering activities including requirements analysis, coding, testing, documentation, repository analysis, and planning. However, the field faces significant challenges in how these practices are managed and governed.

Five Key Research Areas Identified

The paper identifies five research areas requiring attention:

  1. Prompt artifacts and standardization – establishing consistent approaches to prompt creation and management
  2. Evaluation and benchmarking – developing rigorous methods to assess prompt performance
  3. Lifecycle integration – embedding prompt engineering into existing software development workflows
  4. Human-AI collaboration and skills – understanding how teams work with AI systems and what skills are needed
  5. Governance, privacy, and technical debt – addressing the broader implications of prompt engineering adoption

The Challenge of Informal Practices

A core finding is that prompts and related instruction artifacts are often created and evolved through task-specific and informal practices, with limited support for their systematic evaluation, management, traceability, and governance. This suggests significant opportunity for research-driven standardization and tool development across the sector.

Source: arXiv