Reproduce “A Phased Workflow for Operating LLM-Based Coding Agents” 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
Ante Kapetanovic, Tomislav Duricic, Andro Mercep, Emanuel Lacic
The paper proposes a four-phase workflow for agent-assisted software development: research, planning, task definition, and implementation. Practitioners validate findings, curate persistent handoffs, define bounded machine-readable tasks, and use fresh sessions with logs and version control. The proposal reflects reported experience, not an evaluation: it provides no comparison or metric showing improved reliability, productivity, or other outcomes.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.