Welcome to the SageTube blog
· by marcin · meta, announcements
This is the first post on the SageTube blog. It exists for two reasons: to test the blog system itself (this post is also the engineering acceptance test for Phase 2 of our AI visibility initiative) and to lay out who writes here, how, and what to expect.
What this blog covers
SageTube turns YouTube channels into searchable, citable Experts. Every answer the product gives cites the source video and timestamp. The blog covers the same surface — how to use SageTube end-to-end, how the citation grounding works under the hood, how creators can archive a back catalogue as a knowledge base, and how AI search engines actually find and quote the public web.
Two tracks: product education (how to use the Chrome extension, how to build your first Expert, how monetization works) and topical depth (why one billion hours of YouTube is functionally unsearchable, how RAG citations differ from web-search citations, what AI engines reward when they decide whom to cite).
How posts are written
Every post is drafted by an AI agent — typically Claude Code — against SageTube’s own codebase, FAQ, and configuration files. The agent is required to link every product claim to a code reference so the founder reviewing the PR can verify the claim is true. Drafts that make claims unsupported by the codebase get sent back.
A founder reads the PR end-to-end, runs the curation checklist documented in webapp/content/blog/AUTHORING.md, and either merges or sends back. The byline is the founder’s name because the founder is accountable for the accuracy of the post, not the agent that drafted it.
This is deliberate. AI engines have become noticeably more selective about which sources they quote in answers, and low-effort AI-generated content is one of the patterns they down-weight. Every post on this blog is AI-drafted but human-verified — every factual claim, every product feature description, every screenshot is checked against the live product before publishing.
How to subscribe
The blog ships an Atom feed at /blog/rss.xml. Add it to any feed reader. There is no email subscription yet; that will come if the audience grows enough to justify the deliverability work.
What’s next
The first content track is a four-pillar arc on citation grounding, time scarcity in long-form content, the Chrome extension, and creator knowledge bases. The first pillar — Citations, Timestamps, and Trust: How SageTube Answers Are Grounded — is in draft now. Two shorter cluster posts per week will follow.
Thanks for reading the welcome post. The real content starts next week.
Frequently asked questions
- Who writes the SageTube blog?
- Posts are drafted by AI agents using SageTube's own codebase and product documentation as ground truth, then reviewed line-by-line by SageTube's founders before publishing. The founder's name on the byline is the curator who verified the post is accurate and non-misleading.
- How often will new posts land?
- The current cadence is one in-depth pillar post every three weeks and two shorter cluster posts per week. Subscribe to the Atom feed at /blog/rss.xml to get notified.
- Is the content generated by AI?
- Drafts are AI-generated against SageTube's actual code and FAQ; every product claim is checked against a code reference before merge. We do not ship raw AI output. Founders sign off on accuracy and replace any AI-sounding language with plain English.
- Can I republish or quote SageTube blog posts?
- Yes, with attribution and a link back to the original post. Quotations should not exceed 200 words; for longer excerpts please email hello@sagetube.ai.