Engineering Blog — Notion-Backed Publishing

Overview#

A publishing platform for writing about engineering practice, interview preparation, mathematics, and the smaller problems worth recording. It runs on a deliberately unusual content pipeline: Notion is the CMS.

That choice solves the real problem with a personal blog, which is not building it but continuing to write for it. Authoring in Notion means drafting on any device, in a tool already open, with no local repository, no build step, and no markdown files to manage — while the published site remains a fast, statically generated Next.js application rather than a Notion page share.

How It Works#

• Notion as the content database — each post is a database row with typed properties for title, slug, summary, tags, publication state, and date, so the database schema is the content model
• Block rendering — Notion's block tree is mapped to the site's own components, so headings, code blocks, callouts, and images render in the site's typography rather than Notion's
• Static generation with incremental revalidation — pages are prebuilt and served from the edge, then revalidated on an interval, so publishing is a Notion toggle rather than a deploy
• Draft and publish states driven by a database property, so unfinished work never appears
• Tag-based indexes generated from the same source

Features#

• Syntax-highlighted code blocks with language detection
• Reading time and post metadata derived at build time
• Tag filtering across the archive
• Newsletter capture through ConvertKit, integrated into the post flow rather than bolted on as a popup
• Full metadata and social previews generated per post
• Custom domain on a global edge deployment
• Responsive typography tuned for long-form reading rather than inherited from a default

The architecture is the point: separating the authoring experience from the delivery platform means the writing tool can be whatever keeps you writing, while readers still get a static, fast, properly indexed site.