Reproduce “Can LLM Agents Be Forward-Deployed Engineers? Requirement Discovery and Delivery for Enterprise Text-to-SQL” 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
Marianne Menglin Liu, Rongguang Wang, Zheng Wang, Yewen Wang, Leonid Boytsov, Tao Sheng, Sujith Ravi, Dan Roth
The authors study whether LLM agents can discover requirements through dialogue and deliver an enterprise Oracle text-to-SQL system. They evaluate requirement recovery and eight delivery runs against Spider-dev accuracy, production SQL validity, and CPU latency targets. Claude Opus 5.5 averaged 81.7% recovery, while only two runs passed all targets; this bounded case study does not establish general enterprise readiness.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.