A read on where AI fits in Axios Local's stack. And where it doesn't.

Why it matters: The hard part of scaling Axios Local is not the writing. It is the supporting work that has to happen around each reporter so the reporter can stay on the story.

The 60/30/10 read.

• 60% of the work around a local newsletter is traditional code and database work. Subscriber management, ad slot inventory, open-rate reporting, reader-reply routing. Plain software solves this. LLMs here are overkill.
• 30% is rule-based logic. Sponsorship pacing, local ad creative rotation, editorial cadence calendars, new-city launch checklists. Solved by explicit rules that a team can read.
• 10% is genuine AI work. The Axiomizer's draft polishing. Reader-reply summarization at scale. Editorial research augmentation. This is the slice where LLMs earn their place.

The teams that waste budget put an LLM on the 90% that should have been code or rules.

The case: KPMG UK, one of the Big Four. Eduba's founder Jake Van Clief and team ran methodology training for 40+ executives in a regulated, senior-audience environment. A structured read on which decisions at a peer firm belong in which layer. Same shape of conversation.

The paper: Interpretable Context Methodology. Folder structure as agent architecture. Published, ACM TiiS. Relevant to any org trying to standardize how multiple teams run the same playbook. Repo: github.com/RinDig/Interpretable-Context-Methodology-ICM-.

The frame for Miami: Two reporters. 2.7M residents in Miami-Dade. Five sends a week. Arlington sales team switching context 22 ways. Miami is where the capacity math gets tested. It is also the market the 2026 Florida expansion will lean on for whatever playbook Axios Local has codified.

Where Eduba has shown up before.

• Correlation One, Pacific Life, Colgate-Palmolive. 1,500+ enterprise learners trained since May 2025. 6,000 to 9,000 hours saved per year. 95% still using the tools 30 days after the workshop.
• KPMG UK (Big Four). 40+ executives, methodology rather than tooling.
• Feeld. Scoped sprint with the CTO. Workshop, advisory, Organizational Context Architecture.
• VigilOre. 160+ hours of compliance work collapsed to under 5 minutes at the orchestration layer.

Governance note: Eduba also publishes the Ethics Engine, a psychometric assessment tool for evaluating ideological and moral patterns in LLMs. Paper: arxiv.org/abs/2510.11742. Repo: github.com/RinDig/AuditEngine. Useful when the Cox side of the house asks about editorial AI posture.

Partnership note: Eduba partners with NLP Logix for work that sits below the orchestration layer. NLP Logix has been in machine learning since 2011 and runs over 150 data scientists. If a reader-engagement pipeline or a production analytics backend ever sits on the roadmap, that is where it goes.