{"slug": "you-re-not-in-the-loop-anymore-you-re-running-the-loops", "title": "You're Not in the Loop Anymore. You're Running the Loops", "summary": "Kevin Old, writing on Every, reports that using herdr.dev's orchestrator with the compound engineering plugin enabled his team to handle hundred-PR weeks and complete unattended dependency upgrades overnight, marking a shift from human-in-the-loop coding to human-run orchestration of multiple AI agents. Old emphasizes that herdr coordinates mixed agents like Claude Code and Codex without vendor lock-in, while the Every plugin codifies standards so each agent pass compounds knowledge.", "body_md": "- Published on\n\n# You're Not in the Loop Anymore. You're Running the Loops.\n\n- Authors\n- Name\n- Kevin Old\n[@kevinold](https://twitter.com/kevinold)\n\nAI agents can do real engineering work now. That changes the question.\n\nThe question was never how fast one person can code. It's how a whole team gets everything done. Agents remove the typing bottleneck. What's left is coordination and judgment.\n\nI started using [herdr.dev](https://herdr.dev) on Monday. By Tuesday afternoon my team\nwas talking about hundred-PR weeks. After thirty years of scaling SaaS, I have not seen\nleverage like it.\n\n## The loop\n\nAn agent doesn't do one thing and stop. It runs a cycle. Plan, act, check, correct, repeat. Write the failing test, make it pass, run the suite, fix what broke. That cycle is why an agent finishes work instead of handing back a first draft.\n\nFor a while, you sat inside that loop. The agent acted, then waited for you. You were the bottleneck.\n\nThe orchestrator pattern is what happens when you climb out and stand above it.\n\n## What I'm watching\n\nRight now, a dozen agents are working at once. One is clearing failing end-to-end tests. One is driving a PR to green. One is picking apart a review plan before I see it.\n\nI'm not inside any of those loops. I'm watching a board.\n\nherdr is what makes that possible. It looks like a dashboard. It's more than that. It's a layer that sits between the agents and lets them pass work back and forth.\n\n## It's not one vendor's multi-agent mode\n\nThis is the part that matters most.\n\nSome coding agents now ship their own multi-agent features. The tool spins up copies of itself and coordinates them. Useful, but locked in. You're orchestrating one company's agent talking to more of the same company's agent.\n\nherdr orchestrates across agents. Claude Code, Codex, Pi, whatever comes next, all on the same board, under the same rules. You stop picking one assistant to bet on. You route each task to the agent that fits and let herdr keep them moving together.\n\nThe leverage isn't a smarter agent. It's a mixed crew, conducted by one person, with no lock-in.\n\n## The other half: compound engineering\n\nAgents are only as good as the standards they run on. Ours are written down.\n\nEverything runs on [Every](https://every.to)'s\n[compound engineering](https://github.com/EveryInc/compound-engineering-plugin). The\nwhole lifecycle is documented: plan, test, review, ship. And every solved problem gets\ncaptured, so the next run reads it instead of rediscovering it.\n\nThat's the compounding part. Each pass leaves the codebase and the process a little sharper.\n\n## The night it ran without me\n\nWe had dependency upgrades stuck for weeks. Not easy bumps. The kind that need judgment: a breaking change buried in a changelog, a peer version that has to move in lockstep, a lockfile that fights back.\n\nI pointed the orchestrator at them and went to bed.\n\nThe harness held. A crew of agents worked the branches, ran the tests, fixed themselves, moved through the labels, and merged when green. Nobody was awake. I woke up to shipped work.\n\nAn agent writing code stopped being remarkable a while ago. A crew running the full lifecycle to merged, unattended, is the real line.\n\n## This is not vibe coding\n\nIt would be easy to call this magic, or a shortcut around real engineering. It's neither.\n\nLook at the repo. Compounding knowledge. Unit tests. Cypress end-to-end scenarios that drive the real app the way a user would. Pre-push hooks, cost gates, harnesses that refuse broken work.\n\nThe tests still pass. The review still happens. The PR is still green before it merges. The only thing that changed is how many disciplined loops run at once, and how little of my day each one costs.\n\nThis is professional engineering, every guard rail intact, at a scale you could not staff.\n\n## Run it yourself\n\nI packaged the orchestration skill as a plugin. It ships the skill and the scripts, sanitized, following Every's conventions, installable in any coding agent that supports them.\n\n[github.com/kevinold/multi-worker-pm-skill](https://github.com/kevinold/multi-worker-pm-skill)\n\n```\n/plugin marketplace add kevinold/multi-worker-pm-skill\n/plugin install multi-worker-pm\n```\n\nIt leans on Every's compound engineering plugin for the worker loop, where\n`/ce-worktree`\n\nand `/lfg`\n\ncome from.\n\n## What this changes\n\nIf you still measure these tools by autocomplete speed, you're watching the wrong number.\n\nHere's the number that matters. Can the practices you already stand behind, the testing and the review and the knowledge capture and the gates on every merge, run across every stream of work at once, without a person carrying each one by hand? That's a leadership question, not a typing-speed one.\n\nYou're not in the loop anymore. You're running the loops.\n\nThe answer to how a team gets it all done is both halves, together.\n[herdr.dev](https://herdr.dev) for the orchestration.\n[Every's compound engineering](https://github.com/EveryInc/compound-engineering-plugin)\nfor the discipline that makes it worth running.", "url": "https://wpnews.pro/news/you-re-not-in-the-loop-anymore-you-re-running-the-loops", "canonical_source": "https://www.kevinold.com/blog/running-the-loops", "published_at": "2026-09-02 22:22:06+00:00", "updated_at": "2026-09-02 22:52:52.054970+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools", "ai-products"], "entities": ["Kevin Old", "herdr.dev", "Every", "Claude Code", "Codex", "EveryInc"], "alternates": {"html": "https://wpnews.pro/news/you-re-not-in-the-loop-anymore-you-re-running-the-loops", "markdown": "https://wpnews.pro/news/you-re-not-in-the-loop-anymore-you-re-running-the-loops.md", "text": "https://wpnews.pro/news/you-re-not-in-the-loop-anymore-you-re-running-the-loops.txt", "jsonld": "https://wpnews.pro/news/you-re-not-in-the-loop-anymore-you-re-running-the-loops.jsonld"}}