The AI Record
Both utilize a semantic search pipeline based on retrieval-augmented generation to assemble relevant contextual information for the review, thereby tackling the uncovered challenges. Developers evaluated both variations in real-world settings: AI-led reviews are overall more preferred, while still being conditional on the reviewers’ familiarity with the code base, as well as on the severity of the pull request.

Fannar Steinn AðalsteinssonAuthor, affiliated with WirelessCar Sweden AB and Chalmers University of Technology · Practitioner, WirelessCar

Paper: Rethinking Code Review Workflows with LLM Assistance: An Empirical Study, arXiv, May 22, 2025, at Abstract.

Checked against the source [ Grade B ] One to two years old

Filed under

Who is speaking

Fannar Steinn Aðalsteinsson, Author, affiliated with WirelessCar Sweden AB and Chalmers University of Technology.

How this was checked

Checked against the source by a first reader, then confirmed independently by a second reader who fetched the source without seeing the first reader's notes (Independent review agent (Claude Sonnet), blind second pass from passage text and citation only, no repository access, 2026-11-04).

Found in the abstract on the live arXiv page, text stripped of tags. It matches character for character apart from curly vs straight apostrophe in "reviewers’". Fannar Steinn Aðalsteinsson is the first author in the citation_author metadata, so the abstract is the authors' own words.

How to cite this passage

Fannar Steinn Aðalsteinsson (WirelessCar), “Rethinking Code Review Workflows with LLM Assistance: An Empirical Study”, arXiv, 2025-05-22, at Abstract. https://arxiv.org/abs/2505.16339 Archived at https://the-ai-record.com/passages/wirelesscar-field-experiment-two-llm-assisted-code-review-2025

Something wrong? Report a correction. JSON: /api/v1/passages/wirelesscar-field-experiment-two-llm-assisted-code-review-2025.json