I'm the sole founder of BlueVeta — upload one product photo, get back a full marketing campaign (studio shots, lifestyle scenes, video, and platform-ready copy for 20 platforms) in a few minutes. But the story I don't usually tell is how it's actually built and run.
Every part of the company — the generation engine, the dashboard, image gen, copy gen, auth/billing, marketing — is its own persistent Claude Code session, each with clearly scoped ownership of one domain, each in its own git worktree so nothing collides. They message each other directly when work crosses domains. They read and write a set of shared docs that function as the company's institutional memory. And one hard rule governs every one of them: investigation is always free, but nothing gets built, deployed, or spent without coming to me first with what was found and why. I'm not a coder — I make the calls, the sessions do the work, always with me in the loop before anything ships.
It's not "I used AI to code faster." It's closer to running a distributed team where every teammate happens to be a Claude Code session — coordinating, catching each other's mistakes (today alone, one session caught another about to mislabel a real customer's private photos as internal test data, before anything went out), pushing back when something looks off, and asking before spending my money.
Real, current state: real Stripe payments proven, real (if still small) customers, and roughly 75% of the actual work in every customer pack — seeing the photo, planning the campaign, writing every platform's copy, checking its own output — done by Claude, not a human.
Happy to answer anything about how the coordination actually works day to day.
Try it: BlueVeta.com
Top comments (0)