Come Up With An Entirely Novel Way To Develop Software
The Prompt That Named The Whole Thing
A re-think-everything mandate anchored to one shippable proof:
OK, here's our goal- come up with an entirely novel and new way to develop software in the AI agentic age. The broad idea is to take the conceptual model of the Loop 2.1 system and use it to create a way to structure both the development of software, using AI agents governed with managing a portfolio of in-context software, as well as the actual software itself, leaning into a model that looks to the Space Shuttle for fault tolerance and recoverability. It's re-thinking everything. Let's talk this through. The end product, the thing that will mark whether this is a success or failure, is a set of developed guidelines, rules, and files that I can use to develop an app for my father-in-laws deer butcher company in three days. The system we come up with, named the Loop MMT system, must allow me to develop first-class enterprise level software as a solo one man dev leading an AI coding assistant. So our job, again, is to come up with that system. Let's start by you looking over everything and hitting me up with a list of questions.
— Shea Gunther · operator drop, 10.2006
The pattern that recurs across everything after it: an unbounded ambition pinned to one concrete, shippable proof. The mandate is enormous; the success criterion is a single app that either exists in three days or does not.
This is the prompt that named the thing you are reading about. The operator takes the conceptual model of Loop 2.1 — in-context modules, a portfolio governed like software, fault tolerance borrowed from the Space Shuttle — and asks for it to be turned into a way of developing software with an AI assistant, as a solo one-person dev.
What keeps the ambition honest is the last sentence: the whole system will be judged by whether it can produce a working app for a real deer-butcher company in three days. A re-think-everything brief that most people would leave as pure vision is bolted to a falsifiable, dated deliverable — and the prompt closes not with a demand to build, but with a request for the list of questions to start from.
The wider frameThe founding commission: an unbounded mandate made testable by one shippable proof, and opened with questions rather than orders.
“novel and new way to develop software in the AI agentic age” — The mandate — stated at maximum scope, on purpose. Nothing is assumed settled.
“the conceptual model of the Loop 2.1 system” — The seed — the working computer the operator had already built becomes the template for how to build everything else.
“an app for my father-in-law's deer butcher company in three days” — The falsifier — the single concrete test the whole system is measured against. Ambition anchored to a shippable proof.
“you looking over everything and hitting me up with a list of questions” — The opening move — the commission begins with inquiry, not instruction.