The JSONata rewrite story in the existing Grey Haven item points to a practical AI opportunity: eating tooling debt. Migrations, rewrites, adapters, schema cleanup, tests, documentation, and integration glue are expensive because they require context, patience, and careful review. They also block new product work long after the original system stops being strategic.
The Grey Haven read is that tooling debt is a bankable AI wedge because the buyer can already see the cost. This is not a vague labor replacement story. It is a constrained workflow where AI can understand an old system, propose changes, generate tests, and keep humans in review until the migration is safe.
Engineering leaders, platform teams, founders, and private equity operators should care because neglected infrastructure quietly consumes budget and slows product velocity. The opportunity is less like a generic copilot and more like a migration engine for a specific painful workflow.
Operator guidance is to start with bounded debt: one schema, one adapter family, one test generation problem, or one legacy integration path. Track backlog reduction, defect rate, review hours, and outage risk. The watch item is whether tooling-debt products can prove safety, not just speed. Migration work wins when generated changes come with tests, diffs, explanations, and an easy way for reviewers to reject bad suggestions. For this signal, the near-term operating question is whether the team responsible for the real ai opportunity isn’t replacing labor - it’s eating tooling debt can translate the market lesson into one governed workflow, one accountable owner, and one measurable decision point. Source and provenance: manifest record 23, published Grey Haven backfill item referencing the JSONata rewrite story, with no external source URL in the manifest.
