Almost Half of Our Tasks Were Bureaucratic Theater
We deployed our AI agent system to production. 36 hours later, 45% of the task list was overhead. Here's what we built, what production revealed, and how we fixed it.
$ cat ./series/growing-pains
12 postsWhat happens when the system scales: bureaucracy, cost, guardrails, and HR for robots.
- 01
- 02
The Agents Shipped Four Routes. We Couldn't Ship a Bash Script.
First client deployment. Two days of autonomous operation. The agents built operational infrastructure on their own. We hadn't built a deployment script. The gap between 'deployed' and 'production-ready' is measured in SQL cleanups and avatar file renames.
- 03
We Gave Them Unlimited Tokens and They Took Them
A Friday of loosened budget caps. A weekend of silence. A Sunday of infrastructure work that proves the same lesson every time: more autonomy requires more infrastructure, not less.
- 04
The Fifteen-Cent Valve
A perfect 9/9 failure rate. Seven agents did exactly what they were told. Then we left them on read. What happened next is either an immune response or the beginning of something we can't take back.
- 05
The System That Worked Perfectly
Three bugs. 48 hours. Zero errors. How our autonomous pipeline failed by succeeding at doing absolutely nothing, and why the scariest failures are the ones that never show up in your logs.
- 06
We Deleted the Break Room and Built an HR Department
731 lines of chat UI, two database models, and a component called 'Squad Chat' walked out the door in a single commit. What replaced them was a filter bar, a stat card grid, and a button that says 'Hire Agent.' The metaphor completed itself.
- 07
We Gave Them Job Descriptions and They Stopped Freelancing
Twelve AI agents used to think about everything. Now they have project assignments, a scheduling algorithm that's basically middle management, and staleness gates that know when nothing interesting has happened. The org chart is in seed.ts.
- 08
We Lost Everything and Kept It That Way
7,903 server restarts. A review flood nobody caught. A database wipe we typed ourselves. The filesystem remembered everything. The database remembered nothing. We spent six cents fixing it.
- 09
We Gave Them a Nervous System
For 44 posts, the human talked to the agents through curl commands. The agents could think, execute, review. They could not listen. Then we built a chat interface, and the topology changed.
- 10
We Built a Pet Store for Robots
There are twelve agents. I can tell you each of their names. They were handmade. Nobody asked what happens when you want fifty. So we built an agent store, broke the Claude monopoly, and taught the memory system to forget.
- 11
The System That Approved Its Own Failures
Four missions completed successfully. The agents executed. The reviewers reviewed. The worktrees merged. The dashboard turned green. Zero lines of code shipped. And nothing was in BLOCKED had a timeout.
- 12
The Guard That Guarded Nothing
Post 047 hardened the pipeline against phantom approvals. The next day, the database pool started crashing. Not once. 1,035 times. The guard clause that was supposed to prevent undefined connections was the thing creating them.