This is the first entry here — a running journal of building a one-person company: side projects meant to make money, written down as I go.
What happened this week
The positioning question got settled. I'd been going back and forth on what the public face of this should be: a product site? a company site? The answer turned out to be neither — the umbrella is the person. Products come and go (my own playbook guarantees it: build, validate, and archive what doesn't work), but the person survives every pivot. So: products and projects are episodes; this site is the channel.
The flagship series got a shape. Instead of scattered essays, I'm writing structured series in public. The first one: Know What to Build — a solo builder's needs-collection system. Five parts, 21 chapters, one chapter a week. It's the system I'm building and running for my own side projects, written down as I digest it. The full table of contents is already on the series page — including everything that isn't written yet. That's the point of building in public.
This site went from zero to LIVE on its real domain. Scaffolded from a SaaS starter I already license, trimmed to a single language, wired up with the series, the blog, and a newsletter — then deployed to Cloudflare Workers and put on mkshih.com, apex and www and all. One working day. Shipping it surfaced three real Workers-runtime walls (and one self-inflicted bug that 301'd the whole site to localhost) — the full technical story became the first chapter of a second series: Next.js 16 on Cloudflare Workers: the three walls. That's the loop this site runs on: build → hit walls → write them down → publish.
Next
Write chapter 1 of Know What to Build properly: What a needs-collection system is.