I have 230+ posts on LinkedIn and don’t own a single one of them.

That thought hit me harder than it should have. Years of writing, ideas, frameworks, hard-won lessons from building teams and shipping products, all sitting on a platform that could change its algorithm, restrict access, or just disappear my archive behind a login wall tomorrow.

So I built something to fix it.

The idea was simple: crawl my own LinkedIn profile, save every post as a clean markdown file with its images, and publish the whole thing as a personal website I control. The execution was a Playwright browser crawler, a markdown archive organized by date, and a static site generator that turns it all into vanilla HTML.

No framework. No CMS. No dependencies I don’t understand.

The thought process behind it matters more than the tech. I didn’t want a backup, I wanted a living archive. Something I could search, browse, and actually point people to. “Here’s what I think about AI coding tools” shouldn’t require someone to scroll my LinkedIn feed for twenty minutes.

Building it with Claude Code collapsed the timeline from weeks to days. But the interesting part wasn’t the speed. AI didn’t just make building faster. It made building simple things feasible again. You don’t need a team or a SaaS subscription to own your own website to host your content. You need an afternoon, a clear idea of what you want, and just enough engineering judgment to direct the work.

The final outcome is a brutalist little website. Dark theme, no JavaScript frameworks, instant load times. Every post I’ve ever written, searchable and browsable, on a domain I control.

It’s not impressive technology. That’s the whole point.

ps: the whole thing is open source if you want to do the same with your posts.

Media