Evaluation of the pairing of Claude and Codex indicates that the useful pairing is asymmetric: use Claude to review Codex, not the other way around.
Abstract
Developers increasingly use two coding agents together: one writes a draft, and the other reviews it. However, it is not clear whether the pairing is worth its cost and time, or whether the order of the pairing matters. We run a controlled experiment on 116 recent hard and medium lcb tasks with Claude and Codex across six conditions to approximate a software practitioner's workflow: both solo baselines, both cross-model orderings, and both same-model orderings. The reviewer sees the problem and the writer's draft but cannot execute tests, which approximates a code review step. Claude review raises Codex drafts from 71.6% to 89.7% ($p_{BH}=.001$); Codex self review raises them to 84.5% ($p_{BH}=.022$). The reverse direction does not pay off: Codex reviewing Claude drafts drops the pass rate from 91.4% to 82.8% ($p_{BH}=.046$), and Claude self review leaves the 91.4% baseline unchanged. Our evaluation indicates that the useful pairing is asymmetric: use Claude to review Codex, not the other way around.
Large language models now generate code, documentation, and analyses, and are increasingly used to review such output. We ask when a second review by a different model helps. Building on the author's earlier preprints, which varied context, repetition, and role structure within one model, we test model independence in...
Coding agents open pull requests (PRs) that claim to speed up software, but studies of human performance fixes say little about how maintainers respond to such a fix or whether its claim holds. From the 71,677 agent PRs of AIDev v4, a text filter and codebook coding by language models and by the authors select 1,262 pe...
Zhen-Yu Qi, Haotang Li, Jin-Fu Chen et al.· 0 citations
Coding agents increasingly share a workstation while drawing on separate providers and subscription allowances. A second agent can inspect a completed answer, but the call spends another pool and may provide no substantive finding. We describe an advisory cross-provider review contract: distinct resource pools, bounded...
This paper studies 37,623 provenance-labeled pull requests from five commercial agents and combines the AIDev dataset with 58,792 cached GitHub API responses to measure security smells in added code, structural maintainability, post-merge churn, revert rates, and human review behavior.
Aim/Purpose: This study maps the structural tension between human-centric Clean Code conventions and the architectural demands of LLM-based coding agents. It proposes NUDEE, a management framework for governing codebases optimized for both human readability and AI digestibility.
Background: Clean Code prescribes deep...
A controlled experiment comparing five informationally equivalent specification formats across six models from three vendor families finds that structured architecture specifications serve as a capability equalizer, with value inversely proportional to model strength and the largest returns for cost-optimized deploymen...
A. Canedo· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
