AI code fixes overload review and build systems in big projects
Orchestrating AI-Assisted Code Remediation: Socio-Technical Bottlenecks in a Large Industrial Repository
Software Engineering
Summary
Fixing lots of code problems by hand is slow and costly. The authors studied how using AI tools to quickly fix code issues in a large industrial project caused new challenges. These fixes created many code changes that overwhelmed automated building and human review systems. The study found that managing how and when code changes are grouped and reviewed is very important when AI makes editing easy. This helps keep the repair process smooth and sustainable in big software projects.
What this means in practice
- •For software engineering teams: Manage AI-driven code fixes by controlling commit size and review processes to avoid overloading build and review workflows in large projects.
- •For devops teams: Adapt continuous integration systems to handle bursts of AI-generated commits by batching changes and optimizing build resources.
Tested on one dataset.
Authors
Andreas Bexell, Lo Gullstrand Heander, Emma söderberg
Abstract
Background: Code degradation in large, long-lived codebases is costly to remediate through manual refactoring and opportunistic clean-ups. LLM-based coding assistants can perform mechanical remediation at scale, but their impact on industrial workflows is underexplored. Objective: We investigate how massive AI-assisted code remediation affects build-on-commit continuous integration (CI), code review, and team coordination in a large industrial repository, and which socio-technical bottlenecks constrain such remediation when source editing becomes cheap through AI assistance. Method: We report on a 15-day exploratory single-case field study in which an experienced developer used a command-line AI coding buddy to remediate widespread issues in a closed-source industrial C++ repository. We triangulate Gerrit metadata with a developer diary and team chat, analyzed through descriptive statistics and qualitative coding. Results: AI-assisted remediation rapidly generated hundreds of commits touching thousands of lines, saturating CI and reviewer attention. Naïve per-file commits overloaded build-on-commit CI; Switching to directory-based batching and capping the number of files per change restored throughput, but still required explicit review solicitation, negotiation of acceptable commit granularity, and iterative follow-up to resolve build and static-analysis failures. Conclusion: When mechanical editing is cheap, CI capacity, review effort, and change orchestration become primary bottlenecks. Sustainable AI-assisted remediation in very large repositories requires deliberate control of commit, review, and CI batch granularity and treating semantic change sets, such as ``fix all instances of warning X'', as first-class units of work that can be sliced differently for developers, reviewers, and CI.