Reproduce “Agents as Software: A Programming Languages Agenda for Agent Reliability” 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
Shraddha Barke, Adithya Murali
The authors propose a programming-systems agenda for reasoning about AI agents as programs distributed across prompts, tools, memory, policies, and orchestration. It combines pre-run analysis, runtime monitoring of traces and state, and repair informed by failures. In an illustrative refund workflow, a monitor blocks an action lacking required evidence; the essay reports no experiments or measured reliability gains.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.