Notes on product decisions, applied AI, software and data systems, and research-driven content.https://jodyink.com/en-ushttps://jodyink.com/writing/what-a-release-record-should-keep/https://jodyink.com/writing/what-a-release-record-should-keep/The retained release manifest of one studio training build recorded five reasons the build had not been promoted and three things it did not claim. The record is more useful than a clean one would have been.Thu, 10 Sep 2026 00:00:00 GMThttps://jodyink.com/writing/when-deterministic-rules-are-better-than-a-model/https://jodyink.com/writing/when-deterministic-rules-are-better-than-a-model/A four-question test for choosing between explicit rules and model behavior, grounded in data work, reporting analysis, and AI-assisted writing.Thu, 27 Aug 2026 00:00:00 GMThttps://jodyink.com/writing/keeping-evidence-attached-to-ai-assisted-writing/https://jodyink.com/writing/keeping-evidence-attached-to-ai-assisted-writing/How SceneArc checks an AI draft’s references against the notes given to the model, and why the claims still need review.Tue, 21 Jul 2026 00:00:00 GMThttps://jodyink.com/writing/why-a-rewatch-deserves-its-own-history/https://jodyink.com/writing/why-a-rewatch-deserves-its-own-history/A product-modeling decision in CutSeen: repeated viewings become separate runs, so chronology, notes, and uncertainty are preserved instead of flattened.Sat, 04 Jul 2026 00:00:00 GMThttps://jodyink.com/writing/before-replacing-a-legacy-report-reconstruct-what-it-knows/https://jodyink.com/writing/before-replacing-a-legacy-report-reconstruct-what-it-knows/Two reports show the same total, but the new report counts the wrong records. Follow a four-request example to check the records, fix the filter, and verify the result.Fri, 22 May 2026 00:00:00 GMThttps://jodyink.com/writing/a-transcript-is-preparation-not-proof/https://jodyink.com/writing/a-transcript-is-preparation-not-proof/How source records, explicit evidence tiers, and provisional attribution keep AI-assisted transcript work useful without granting it false authority.Sat, 02 May 2026 00:00:00 GMThttps://jodyink.com/writing/rejecting-a-model-is-a-useful-result/https://jodyink.com/writing/rejecting-a-model-is-a-useful-result/What a small, task-specific evaluation revealed about source grounding, context pressure, and the value of a clear no-go decision.Tue, 10 Mar 2026 00:00:00 GMThttps://jodyink.com/writing/what-a-public-case-study-can-honestly-claim/https://jodyink.com/writing/what-a-public-case-study-can-honestly-claim/A practical way to separate relationship, implementation, availability, and outcome before portfolio copy turns evidence into implication.Thu, 26 Feb 2026 00:00:00 GMT