· XingAI Tech Blog
The Archive Was Already Written. It Just Was Not a Site.
What was true
This repo already had 88 English + 88 中文 posts. Last write: 2026-08-05. Public GitHub Markdown.
blog.xingai.app had no DNS. No generator. No robots.txt. For a crawler, the archive did not exist.
What shipped in the repo (0.1.0)
Next.js at the repo root reads /posts at build time and emits HTML. Markdown stays the source of truth. There is still no CMS.
| Surface | Behavior |
|---|---|
| UI | en / 中文 / 한국어 |
| Article bodies | English + 中文. Korean UI shows a note until KO translations exist |
| Indexing | /robots.txt allow-all, /sitemap.xml, /llms.txt, FAQ JSON-LD |
| Legal | /legal/privacy, /legal/terms, /legal/disclaimer |
Local production build (2026-09-20): 869 static routes. Sitemap: 864 URLs. Post bodies are in the HTML, not a client shell.
Deploy notes: VERCEL.md. xingai.app now lists the blog and links it from /engineering.
What is not true yet
blog.xingai.app is still not on the public internet. The site runs locally (npm run dev → port 3010). Production needs a Vercel project and a DNS CNAME for blog. Until that lands, do not treat the domain as live.
This archive is engineering notes for builders. It is not an investing blog and not a traffic funnel for Invest AI.
Takeaway
Ship the files you already wrote. Do not start a second writing pile because GitHub Markdown is invisible to search.