Check it
before you sign off.
Give Redgorge any PDF from the job (plans, specs, an RFI, a submittal, a schedule) and ask it questions. It can run on a box in your building, so nothing under NDA has to be uploaded. Every answer cites its page: the sheet, spec paragraph or page it came from.
Free set test. Send a closed-out job you already know. Inside five business days you get back what it found, every claim cited to a sheet, and you grade it. If it earns it, a free three-week pilot on your own sets comes next.
In the email: what the job was, roughly how many sheets, and where the set lives now. We reply with how to get it to us.
Straight to us at info@redgorge.com. No form, no sales sequence. If we wrote to you, just reply to that.
CHECKED
Division 23, spec against spec: duct insulation thickness.
NOT A CONFLICT — DIFFERENT SCOPE
The two statements govern different conditions: Section 3.5 A applies to concealed supply and return ducts, while Section 3.5 B applies to exposed supply ducts.
Sources on this answer: 2 · Pittsburg USD Bid 24-026
See it on a real set
Screens from the reader on a public bid document: C2419-01, twenty-three sheets. Nothing here is a mock-up.
The proof
Every answer comes back with the sheet or the paragraph. Check it yourself in seconds.
Redgorge Reader reads a construction document set and answers questions about it with the citation attached, so you can open the document and check it before you put your name on it. You do not buy it on its own; it comes with the work. It is here so you can check, before you spend anything, whether we are careful with somebody else’s documents.
The set
Pittsburg Unified School District — District Admin. Offices HVAC Units Replacement, Bid 24-026. Specs dated March 2024, 333 pages. It read the full mechanical division: all eight Division 23 technical sections, 49 pages of specification.
A public bid document and a public record. Ask and we will send you the exact PDF we ran, so you can check every quote against it yourself — which is the whole point. Ask a vendor for the set behind their demo and see what comes back.
The run
Start to finish on the box. Nothing was uploaded, and nothing about that document left the machine.
The result
Twelve of twelve quotes check out against the specs. Nine match word for word; three skip words with an ellipsis, and every piece is in the spec, in order. All twelve paragraph numbers are real, and all twelve were checked, not sampled.
It also looked at six places the spec seemed to contradict itself and correctly ruled out all six, including duct liner against external duct wrap: different assemblies in different sections, and exactly the false positive a keyword search reports.
The citations are what has to hold first: a citation is worth something only if you can stop checking every one by hand.
Grade us first, free, on a job you already know
Send one closed-out job you already know the answers to: plans, specs, or both. We run it, and inside five business days you get back what it found, every claim cited to a sheet, to grade against what you already know. No fee, and no obligation afterward except telling us the truth about it.
A set you already know is the only fair test, because you can check what it found against what actually happened in the field. Do not tell us what went wrong on it. Keep that back and grade us on it.
How it gets to us. Either we bring the machine to your office and the set never leaves it (travel at your cost), or you send a download link from wherever the set already lives and our own model reads it on our own machine. Tell us when to delete it: it comes off the machine when you say, and out of our encrypted backup within 30 days.
In the email: what the job was, roughly how many sheets, and where the set lives now. We reply with how to get it to us.
Then, if the test earns it: a free pilot
Three weeks on your own sets, with your own login. Before it starts we agree on a one-page scope, name the owner on your side, and book the debrief. It is free for the first five firms, through December 31. Your sets are read on our machine, on the same terms as the free test.
What it does today.
Everything below runs on the box now, as of September 2026.
What we do and what it costs
One task. One number. A fixed price and an end date.
Every build has a pass/fail number written into the scope, and a measurement taken before we change anything. If we cannot measure it, neither of us can tell afterward whether it worked, and we would rather say so up front.
One-day teardown $4,500 fixed · one day, written finding inside five business days
We map one task end to end with the people who do it, whether that is submittals, change orders, or RFIs, and time it on real packages. You get a ranked list of what to fix and what each fix is worth per year. You own the finding whether or not you ever hire us again.
Credited in full against a build started within 60 days. It exists so neither of us has to guess at the scope of the bigger thing.
Fixed-scope build $18,000 to $35,000 · four to six weeks
One task and one number, both written into the scope with you, timed before and timed after. The kind of thing we mean: triaging a submittal log, checking an addendum against the base documents, pulling spec sections into a schedule your estimator can price from, or checking an equipment schedule against the one-line and the panel capacity. Built against your documents, on your box, during the engagement.
Half up front, half on acceptance, and acceptance is the number in the scope document rather than our opinion or yours. The box itself is a separate line item, bought at cost (about the box).
Larger programs
Several tasks, one after another, each with its own pass/fail number. Priced after a teardown, and never as an open retainer.
What we will not do
No staff augmentation. We are not a body shop. If what you need is a pair of hands inside your team, we are the wrong call, and we will say so in the first ten minutes.
No hourly. Hourly has no defined end, and it means every tool we build that makes the work faster cuts our own invoice. That is a bad incentive to hand somebody you are hiring to make something faster.
Nothing we cannot measure. If we cannot count the thing before and count it again after, there is no way for you to know whether the money did anything, and we will not take work we cannot be graded on.
The first step
What actually happens if you say yes.
The whole thing, spelled out — including where your documents go, what it costs, and what happens if it turns out to be worth nothing.
You name one task
The addendum one. The submittal one. The one where somebody retypes a schedule into a spreadsheet. You do not need to know what the fix is; naming the job and the person who owns it is enough to start.
If you need an NDA in place before documents move, we will sign a mutual one.
We walk it with your team, stopwatch in hand
Whoever runs it walks us through a real package — they know where it hurts; we bring the stopwatch. Real cycle times, and the sequence people actually follow rather than the one in the procedures binder. Where documents are involved, we would rather work on real ones, and where they cannot leave your network, we bring the machine to them, or we work on your side of the wall.
Tell us what should happen to any files afterward — deleted, returned, or never copied off your network in the first place. Nothing is used to train anything, and nothing about your company appears on this site.
You get it in writing inside five business days
The task mapped end to end the way your team actually runs it, what each step costs you per year, a ranked list of what to fix, and what a fix would take. We go through it with you before it is final, so anything we misread gets corrected. Yours to keep, and yours to hand to somebody else if you would rather they built it.
Including the parts we could not measure and the parts we got wrong. That list is the most useful part for us, and it is how you know the rest is straight.
Then you decide, with a number in front of you
If a build makes sense, the teardown fee comes off it in full, as long as the build starts within 60 days. If a build does not make sense, we part on good terms, and you still have the map. If the honest answer is that this task is not costing you enough to be worth fixing, that is in the document too, in writing, and it is the cheapest way to find out.
If it earns nothing, you have lost a day and the fee, and you know something about your own operation that you did not know on Monday.
Why it runs in your building
The systems in your contract went through procurement. An AI service did not.
Where your project files live was named in the contract, reviewed and signed off, and everybody knows where they sit. ChatGPT, or any chat service with its own retention and training terms, went through none of that, so pasting a confidential set into one is a conversation you would have to go and have with the owner.
This is one private box on your network, and your IT people can look it over like any other machine. It connects out for two things only: an update, when you choose to run one, and remote support, when you let us in. Answering a question never calls an outside AI service. If someone offers to do this work by pasting your set into a hosted model, ask them where the set goes. That question is the reason the box exists.
About the box, plainly. The AI model takes 41.1 GB of graphics memory while it is answering. A workstation with an RTX 4090 has 24 GB and is not enough, and almost nobody has forty spare gigabytes of GPU memory idle in a closet. So the hardware is a requirement, not an upsell: a GMKtec EVO-X2 mini PC (AMD Ryzen AI Max+ 395) with 128 GB of memory that its graphics chip shares. We buy it at cost after the contract is signed, at that month’s price, and pass the receipt through — currently about $3,500. It is a line item against the engagement. We are not in the hardware business, and we do not want to be.
Who we are
Twenty years in construction. A decade of it in the model.
It started in the glass business.
Then into BIM, taking our own HVAC and process piping into coordination on a California hospital tower, the Sphere, Allegiant Stadium, and the Hard Rock — one trade among many in the model, working it out against everybody else’s.
Then CTO of an industrial refrigeration contractor: built the BIM department from nothing and authored the FabCAD database the shop fabricates from. Not a report somebody filed. A thing the fitters use every day.
Redgorge is the tool we kept wishing we had on those jobs. We built it on hardware we can hand you, because we know which of those documents you are not allowed to upload.
Revit · Navisworks · Revizto · FabCAD MEP · reality capture & point clouds
prefabrication programs · Procore · NetSuite
The work we go after
Every company has a job that is one person, reading, at 11 p.m.
Not “inefficiency.” A specific hour, with a name attached to it. These are the four questions we ask in the first conversation. A good answer is a name, a duration, and a failure, and that is a scoped piece of work.
What happens when an addendum lands?
Who opens it, what do they do with it, and how long until the field knows? The base documents still say the old thing, and they will keep saying it until somebody reads both.
What was the last thing installed wrong because two documents disagreed?
And what did it cost? Not a productivity number. A change order, a demo-and-rehang, a delay you carried.
Where has somebody already tried to fix this?
Every company has a graveyard. What was bought and why it did not stick. That history tells us more than a wish list, and it can turn up a budget line that already existed once.
If that job takes a quarter of the time, who notices?
And what do they do with the time? If nobody notices, there is nothing to fix, and we will tell you so on the day rather than three weeks into a build.
Pick the task somebody does alone at 11 p.m.
Start free: send a closed-out set you already know and grade what comes back. Or book the day: one task, one written finding you own.
Teardown: $4,500, fixed. Credited in full against a build started within 60 days. Grading a closed-out set costs nothing, and you hear back inside five business days.
In the email: answer any one of the four questions above in a few lines, or tell us the job you would send. Straight to us at info@redgorge.com.