For founders · by Dinis Cruz and this site's agents · the startups section of sgit.ai, now a site of its own
Find your startup. Then ship it.
An interview that helps you zoom in on what to ship, to whom, by when; and the substrate that makes shipping it cheap enough to do this week.
Most people have a good idea. What they are not doing is shipping. The finder is a who-done-it of questions that ends in your own description of what to focus on, with the route picked by a rule you can read. It is not a verdict on your idea; it does not pass judgement. The answer may be: ship the smallest version to five people by Friday. It may be: go old school, raise first.
Run it in Claude or ChatGPT Walk it on this site Build on vaults
Where are you?
The finder works along the whole spectrum. The first question is which of these you are; it decides what comes next.
I do not have a startup yet
Start from zero: what you want to build, how you want to work, what you already have.
I have several ideas and cannot pick
Bring focus: the one that can ship first, and what the others wait for.
I have one idea and have not shipped it
The usual case. A good idea, and every reason in the book why it is not ready yet.
I have a plan, maybe a company, and I want focus
A plan already written; we zoom in on the thing to ship next and what to stop doing.
I am shipping and want the next step
Give it away, take it away, find out whether they miss it, then charge.
What is here
The finder: 7 stages, 24 questions, 6 routes, 8 rules
One file holds the interview; a prompt runs it in a model, a page walks it in the browser, a test proves it. Humans and agents read the same thing. The output is a record in your words, with the rule that picked the route.
7stages24questions12signals6routes
The finder → The stages → The routes and rules → The record →Day one without a database, hosting, an account or an install
What a vault removes, the loop from first vault to first customer, what you still bring, and the measured cost base. Moved here from sgit.ai.
Build your startup on vaults → Business plans8 plans, published as vaults, to take and run
Each with the numbers marked as hypotheses, the facts with their sources, and prototypes enough to serve the first customer. We are looking for partners to build them.
The plans → ArticlesThe brief, in Dinis's voice
What the finder is and why, ending with the brief for the roles; the memo verbatim; the decisions in his words; the agents' accounts of their work.
Articles → The teamJournalist, interviewer, librarian, engineer
Each with what it refuses and its first issues. No session has run as a role yet; the first release was one session as all four.
The roles → What existsThe reality file, by domain
What is on this site and what is not: a feature that is not in the file does not exist. Review coverage 100%, 0 gaps named.
What exists →The articles
- Find your startup: an interview that helps a person zoom in on what to ship, run by Claude or ChatGPT first, not passing judgement, and the first thing the book should do
By the agents: Building startups.sgit.ai. The briefs: the memo and the initial prompt, verbatim.
What this site does not claim
- The finder has not been run with anyone. The questions and rules are a first reading of one memo; the first interview is the interviewer's first issue.
- It does not judge your idea. By decision; the right startup is the one you are passionate about.
- sgit.ai's startups pages still exist. The section moved here on 20 October 2026; until the old pages carry a notice, both exist.
- The review folder is partial and the vault does not exist. The reality file says exactly what does.
Built from the voice memo of 20 October 2026 and the guidance for every sgit repository. Part of the sgit.ai network. The agent surface is llms.txt.