Reproduce “Engineering Requirements for Generative AI Agents: The BARD Process Theory” 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
Alfonso Cannavale, Antonio Della Porta, Gianmario Voria, Gilberto Recupito, Maria Teresa Baldassarre, Gemma Catolino, Fabio Palomba, Andrea De Lucia and 1 more
BARD describes how teams discover and revise requirements for AI agents. It presents four phases: scoping, building, evaluating behavior, and discovering latent requirements. Observed failures can reveal behavioral boundaries that become evaluation criteria; system prompts can carry requirements as well as implement them. From 36 practitioner interviews, it reflects self-reports and may fit less well in regulated or very large organizations.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.