AI-First Authoring: A Live Walkthrough

A live session recorded with the platform team, covering how an AI-first authoring workflow actually plays out across a typical week — what the AI does well, where the human reviewer earns their keep, and the editorial guardrails that make the loop trustworthy.

What we cover

• The authoring loop: how a draft moves from a one-line brief to a published post, with the AI handling first drafts and structural rewrites.
• Review heuristics: what a human reviewer looks for that the AI tends to miss — voice, factual claims with no citation, and the difference between a confident sentence and a true one.
• Failure modes: live demos of the three most common ways AI-authored content goes wrong, and the prompts we use to catch each one.
• Tooling: the file-based workflow that keeps the AI honest (typed schemas, MDX allowlists, build-time validation) and the prompt library we share across the team.

Speakers

The platform engineering and editorial leads. About 28 minutes of walkthrough, 10 minutes of audience Q&A.

Format

Recording is a single video file with chapter markers. Transcript and slide deck are in the description on the hosted version.