Lintel

Three ways in

Interested in helpingsolve the housing crisis?

Nothing here is proven yet, and this project takes on one part of the problem: it takes roughly 32 months to deliver an American home, eight of them spent building it, and the two years around the build are information processing. Written by John Mullin. I trained in computer science and mathematics and spent the last ten years building startups; this project reads American building regulation the way an engineer reads an uncompiled codebase. That lens has a limit, and this page is the limit: three questions I cannot answer alone, each with the first thing I would ask for.

Engineer

Licensed, on being professional of record 1 / 3

The instrument requires a determination somebody signs. I do not know what that signature commits a licensee to, or whether the standard of care reaches a machine-generated finding at all.

First deliverable Thirty minutes on what signing would require of you. Not a commitment to sign anything.

Untested → Email j@mullincap.com

E&O underwriter

A broker or underwriter, on what would have to exist before a quote 2 / 3

Florida already mandates rated professional liability for private plan reviewers, so the line is not hypothetical — but a determination reads as a certification of fact, and certifications sit close to the express-warranty exclusion.

First deliverable The list of what is missing. I expect that list to be long, and it is more useful to me than an encouraging answer.

Untested → Email j@mullincap.com

Building official

In a third-party-review state, on a draft amendment 3 / 3

The ask is statutory and small: recognition of an insured determination under provisions that already exist.

First deliverable Redlines on the draft. Marking it up is the useful act; telling me it is unworkable is a finding I will publish.

Untested → Email j@mullincap.com

What you get. Named credit in the evidence register against whatever row your answer fills, unless you ask not to be named. The Gate 0 result before it is published, including if it comes out against this thesis. And the exchange itself published — what you said, what it changed, and where I was wrong.

Nothing is being raised and there is nothing to sign up for. No gate has been cleared. If you think the whole thing is wrong, that is also a useful twenty minutes, and the corrections log is where it would end up.