Code review has been the primary quality gate in software development since Fagan formalised code inspection in 1976. For five decades, having a human examine and comment on a colleague's changes before merge has been a cornerstone practice at organisations of every size. Coding agents are large language model (LLM)-based autonomous systems capable of reading, writing, testing, and repairing software. We argue that coding agents have crossed a threshold of capability at which traditional human code review is no longer a necessary component of a software quality pipeline. Our argument rests on two claims: every stated goal of code review can be served by agents at lower cost and higher throughput; the naive integration in which agents write code and humans remain the mandatory reviewers is a dead end because it neither provides meaningful assurance nor scales with AI-assisted throughput.
I read the paper. It asserts a hell of a lot. For evidence it points only to other papers that do not IMO make claims as solid as are taken by these authors.
I suspect with the right mix of skills, tools, and orchestration, a frontier level LLM could review code as well as human. But this handwavey paper makes no such detailed architectural description and provides scant data for such conclusions.
In short this paper is hypetrash, quite apart from the state of the technology