Article

Specification-first convergence with an AI coding agent: a case study of dismantling a core architectural invariant across 189 files in a 717k-line codebase with no test oracle and no human code review

2026-08-25

Abstract excerpt

<title>Abstract</title> <p>This paper reports a single, fully instrumented case study of a large-scale architectural refactoring by an AI coding agent under a specification-first protocol, with no human review of the generated code and no pre-existing oracle to validate the target behaviour. The task, dismantling a central invariant across a large interdependent codebase, was assessed by the author as effectively...

Topics

Open a Topic to create a Post that cites this publication.

Genomics and Rare Diseases

Identifiers and source

Literature Corpus work: 0c8c11d7-0f59-5c58-b6a4-14443207f756
DOI: 10.21203/rs.3.rs-10800353/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Specification-first convergence with an AI coding agent: a case study of dismantling a core architectural invariant across 189 files in a 717k-line codebase with no test oracle and no human code reviewDOI 10.21203/rs.3.rs-10800353/v1Read

From Requirements to Production: Governing AI-Assisted Software Delivery through a Canonical Requirements ModelSemantic proximity Read

Code is Cheap, Judgment is Not: The Rise of Integration Skill in the Age of AI-Written CodeSemantic proximity Read

Mutation Testing of Task-Scoped State Oracles in Software-Agent Benchmarks: A Cross-Benchmark Empirical StudySemantic proximity Read

Who Broke the Pipeline? Traceability andAccountability in Multi-Agent LLM SoftwareEngineering PipelinesSemantic proximity Read

From Prompt to Production: A Case Study of Human-AI Collaborative Software Development Using Claude Code Multi-Agent OrchestrationSemantic proximity Read

Execution Grounded Multiagent Systems for Reliable Backend Code Generation with Large Language Models'Semantic proximity Read

Intent-Driven Code Synthesis: Redefining Software Development with TransformersSemantic proximity Read

Autonomous AI and Agentic Testing Agents: A Multi-Agent Architecture for Self-Directed Software Quality AssuranceSemantic proximity Read

The World Is a Hypothesis: Epistemic Control for Executable World-Model Agents in Novel Interactive EnvironmentsSemantic proximity Read

Self-Debugging AI: A Comprehensive Analysis of Claude 4.1 Sonnet's Code Generation and Error Resolution CapabilitiesSemantic proximity Read

Artificial Intelligence versus Software Engineers: An Evidence-Based Assessment Focusing on Non-Functional RequirementsSemantic proximity Read

A Systematic Literature Review on Generative AI in Software Engineering: Code Generation and RefactoringSemantic proximity Read