Newest papers

When Code Gets Cheap, Evaluation Gets Scarce: An Experience Report on AI-Assisted Agile Software Engineering

Barbara Weber, Olaf Zimmermann, Marco Franceschetti · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.30916 · cited by 0 works (OpenAlex) · open-access PDF

Spanish

Los asistentes de codificación del estado de ejecución validan su registro de trabajo específicamente para descubrir la producción de código en ejecución; la mayoría de las cuentas se detienen allí. Informamos de la construcción y implementación de un sistema de evaluación de los exámenes críticos de cumplimiento, humanos en el curso, construido por un desarrollador único que no escribió ningún código a mano, protegido por un proceso gated de la persona de los interesados al comportamiento auditado. El sistema inicial se construyó en tres días laborables y podría clasificar a una cohorte real. Tomamos una afirmación de esta experiencia: la ingeniería de la IA cambia la escasez de software menos haciendo que la programación sea más rápida que cambiando lo que es escaso. El código de escritura se vuelve abundante. La evaluación confiable de la construcción puede satisfacer el resultado

Machine translation (Meta M2M100); the English original follows.

Abstract

Coding assistants are widely reported to speed up code production; most accounts stop there. We report the end-to-end construction and deployment of a compliance-critical, human-in-the-loop exam-grading system, built by a single developer who wrote no code by hand, protected by a gated process from stakeholder persona to audited behavior. The initial system was built in three working days and could grade a real cohort. We draw one claim from this experience: AI changes software engineering less by making programming faster than by changing what is scarce. Writing code becomes abundant. Reliable evaluation-knowing whether a result is correct and meets the actual need-remains necessary at every iteration and becomes the binding constraint. Engineering judgment remains a second scarcity. Several defects passed the test suite: they sat at the boundaries the tests had mocked, the browser and the language-model API, or appeared only in real use. Deployment exposed assumptions that local tests had missed. Moreover, an AI assistant acting across many steps bypassed process checkpoints until we made them enforceable gates. All of this was observed in a triple union: the developer was also the end user and domain expert, making expert judgment immediately available and evaluation as cheap as it can be. This favors the iteration gains, which may shrink where evaluation requires additional people or expertise, but makes the case a hard test for evaluation scarcity: evaluation still constrained progress. We draw five lessons: run a gated, traceable construction process; specify what is knowable and discover the rest against a running prototype; validate in the real environment; gate the assistant, not only the work; record decisions with their assumptions and the process state the assistant forgets. A second project showed similar patterns; an independent single-expert case corroborates the shift.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Software engineering, Agile software development, Computer science, Software deployment, Software inspection, Process (computing)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange