2026-10-11 18:01 UTC

Independent use will determine whether Otlet can make local LLM inference practical inside Postgres by reliably managing model jobs, outputs, receipts, and reviewed database writes through background workers.

state: expiredheat: lowuncertainty: highnovelscott: nonelocal-inference postgres ai-workflowsJosh Meek

What is this?

Otlet is described as a Postgres extension, attributed in the case to Josh Meek, that runs local LLM inference through a background worker while Postgres stores jobs, inputs, outputs, and receipts. Its proposed workflow includes reviewed database writes, making Postgres both the system of record and the control plane for asynchronous AI work. The supplied web results establish adjacent approaches for invoking local models from PostgreSQL and emphasize that local inference remains constrained by RAM, compute, and hardware configuration, but they do not independently establish Otlet’s reliability or real-world adoption.

Why it matters to Scott

No intersection found: neither Scott’s wikis nor the radar supplied any pages connecting his positions or active projects to Postgres-managed local inference, Otlet, or this development.
queries asked of Scott's wikis
  • Postgres as AI workflow control plane
  • durable agent jobs receipts and provenance
  • human-reviewed LLM database writes
  • local inference inside application infrastructure
  • background workers for asynchronous AI tasks
  • database-native AI workflow patterns

Measured heat

no measured readings yet — the hourly heat pass fills this in

How the heat travelled

no chain yet — the hourly chain pass fills this in

Evidence (2) — ⭐ canonical anchor

sourceobjectauthorscorecomments
🟧 hnShow HN: Otlet – Local LLM inference "inside" Postgresjosmek11
🟧 echo.github ⭐Otlet is a Postgres extension that runs local LLM inference in a background worker while Postgres stores jobs, inputs, outputs, and receiptsJosh Meek——

Interpretation history

Decision trace