The Loop, AI Operator Course #3
“I put 20 good tasks somewhere. Tomorrow morning there are three PRs waiting for me.”
Three, not twenty. The queue holds the other seventeen until you've cleared these.
The Walkthrough
The whole system running end to end, on a real machine.
Before / After
Before
Developer → AI → code. You're the operator. You load the context every time, for every task, from scratch.
After
Backlog → readiness check → AI developer → PR → human review. You're the reviewer.
The Four Levels
L0: No AI in the loop.
Every line typed by hand. AI is a search engine you occasionally paste into.
L1: AI as a chat window.
You open a session, paste context, wait, review, paste the next thing. One task at a time, and you're in the room for all of it.
L2: AI as a queue (you're here after this course).
A task file clears a context readiness gate, an agent picks it up unattended, and a PR is waiting when you next check.
L3: Multi-agent orchestration.
Agents managing agents. You don't need this yet, and most teams never will. Learn Level 2 first.
The Readiness Gate at 85%
An agent working from a half-specified task guesses, and a wrong guess at 2am is a PR you have to unwind at 9am. The Readiness Gate scores every task file before an agent is allowed to touch it. Nothing below 85% context completion gets picked up.
A gate in front of the model buys you more than a bigger model does.
Cadence: One an Hour, Twice
It's two fires, not one: the readiness gate runs at :30 past the hour, the build runs at the top of the next hour. Write a task before :30 and it's scored within the half hour; clear the gate and it's built and opened as a PR by :00. Best case is about 30 minutes from writing a task to a PR open, worst case about 90.
Neither fires more than once an hour, because that's one human's review capacity. Faster doesn't help. PRs you haven't reviewed pile up the same way the backlog did.
Nothing to Sign Up For Besides This (Or Move To Where You Are Already)
The queue is markdown files. One file per task, in docs/tasks/ inside the repo it belongs to. No tracker account, no API key, no third-party service that can be down at 3am — and the file's location is what tells the agent which repo the task is for, so there is nothing to map or keep in sync.
That queue has exactly one cost: it lives on your machine. When you get tired of losing the ideas you had away from the desk, the port-my-loop skill moves the whole thing to Linear — same rubric, same gate, same cadence, but now you add to the queue from your phone on the bus: an issue filed before 8:30 gets gated at 8:30 and built at 9:00, before you've opened the laptop. That's a bonus module, not a prerequisite.
Running Right Now
2,347 runs
806 hours of unattended agent work, on the same loop this course teaches. No dollar figure attached, because I'd be guessing what your hour is worth.
What You Get
- Install from Claude Code: paste one prompt into a new session and Claude installs the loop with you. No download needed; a zip is there if you'd rather install by hand.
- The install session: about 45 minutes, walks your machine through config, skills and the two hourly loops, then fires both so you watch them work. After that it runs on its own.
- Five Claude skills that run the loop end to end, including port-my-loop for when you outgrow the markdown queue.
- Talk to the course: a personal Course Key and the ask-the-course skill, so your Claude Code answers loop questions straight from the lessons and reference pages.
- Fifteen reference pages for what to read before something breaks, and what to read after.
- The Loop Accountability Check, so every session you run by hand tells you whether it should have been a queued task instead.
- Branch-janitor, as a bonus: audits and cleans up stale branches, no extra setup.
Who It's For
Qualifies: an AI-assisted (L1) developer already using an agent day-to-day, comfortable with git and PRs, who wants a queue instead of a chat window.
Price
Founding (first 5) (current)
€29
4 seats left
Early (next 5)
€49
5 seats left
Standard
€99
Open
Founding price is base only. A testimonial is required within 14 days of your first merged agent-authored PR. One-time payment, lifetime access, and the loop is yours to keep either way.
Add-ons, never bundled at founding tiers: Install With Peter (€450, for companies who'd rather hand the whole session over) and Keep My Loop Alive (€19/mo for my updates as I make them. Cancel and your loop keeps running).
The Guarantee
Your first agent-authored PR within 7 days, or a full refund and you keep the skills. One condition: complete the ~45-minute install session. No other conditions.