SOURCE-LINKED INTELLIGENCE

LLM Post-Training as Brownfield Maintenance: An Industrial Perspective on Dataware Engineering

arXiv · AI, language, vision and robotics · article · Aug 31, 2026 · UTC

Industrial post-training is a brownfield regime. Teams inherit a deployed checkpoint and must land targeted improvements under fixed compute and mixture budgets without regressing the rest. The maintained artifact is increasingly dataware: behavior governed by a curated post-training mixture, updated via bounded mixture patches rather than clean-slate retraining. From an industrial code-generation improvement effort, we offer a maintainer's perspective on why this work is hard in practice, distilling three recurring challenges, zero-sum mixture design, yield as the binding metric, and end-to-e

Read original source ↗ Open in workspace

recordType

paper

region

Global