Reproduce “Agentsway -- Software Development Methodology for AI Agents-based Teams” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Eranga Bandara, Ross Gore, Xueping Liang, Sachini Rajapakse, Isurunima Kularathne, Pramoda Karunarathna, Peter Foytik, Sachin Shetty and 7 more
Old Dominion University and collaborators introduced Agentsway, a methodology enabling AI agent-based teams to manage the software development lifecycle under human orchestration, and demonstrated its application in automating legal case handling with a planning agent achieving an average rating of 4.7/5 for coherence.
Key insights
- Effective software development in the agentic era requires a redefinition of human-AI collaboration, positioning humans as orchestrators and delegating specialized functions to autonomous agents.
- Continuous, feedback-driven fine-tuning of underlying Large Language Models is essential for building adaptive, self-improving AI-native software development systems.
- Responsible AI principles, including ethical oversight and privacy-by-design, can be systematically embedded into agentic workflows through orchestrated multi-LLM reasoning and secure environments.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.