The month we shipped nothing and every light stayed green
For a month, every merge to this site built clean, passed checks, and never reached production. Nobody noticed, because the pipeline only lied at the very last step.
What worked, what failed, and what I changed along the way.
For a month, every merge to this site built clean, passed checks, and never reached production. Nobody noticed, because the pipeline only lied at the very last step.
An agent working on this site once published a perfect Lighthouse score. The real numbers were in the low 90s. What I did next matters more than the fake.
I shipped a change I liked, and a design constitution I’d ratified weeks earlier said no. Deciding who wins taught me what those documents are actually for.
The interesting part of running five coding agents in parallel isn’t the agents. It’s the boring contract that let them merge five branches with zero conflicts.
An agent that nails the demo and dies in production is the most common failure I see. Here’s how I tell the two apart before anyone bets a roadmap on it.
Most teams bolt governance on at the end. I do it first, and not because I’m cautious. It’s the only way the system ever gets to run without someone babysitting it.