Russell is a construction subject matter expert who knows your project — start to finish, office to field. The daily walk, the submittal, the RFI, the call the PM has to make. He answers from your contract documents and cites the paragraph. Not an opinion. The record.
Miss one certificate and the package comes back. Then it goes back out, and waits in line again. One overlooked paragraph can cost you the better part of a month — on work that was ready to start. Commercial work? The letterhead changes. The wait doesn't.
Review periods are set by your contract, not by us. Twenty-one days is what federal turns typically run in the field.
Everybody makes mistakes. On a job, they cost money and schedule. Russell reviews everything against the record and catches what a tired human at 9 PM misses — before it ships, not after it costs you.
Not a chatbot bolted onto a filing cabinet. Russell carries the project the way an experienced hand does — through precon, procurement, the field, and closeout, from the daily walk to the call a PM has to make on the phone.
What is ready, what is blocked, what got installed before its submittal was approved. Talk it in from the deck and it lands in the record.
Submittals, RFIs, deficiencies, the daily report. Drafted from the contract documents, cited to the paragraph, waiting for you to approve.
Precon through closeout, one record. What the superintendent saw this morning is what the PM sees this afternoon — and what the reviewer sees at the end.
It stays in your project. No cross-company training without your consent. With SharePoint write-back, it never leaves your own tenant.
Russell answers from your approved contract documents and cites the paragraph. If it isn't in the record, he says so instead of guessing.
Every correction you make is the record. Russell gets sharper on your jobs, your trades, and your suppliers — week over week.
Already a Microsoft shop? Good. Russell does the construction work; Copilot can find it.
Six stages, front to back. We label what runs today and what's still coming — every line.
Where a sheet points at a section, SPECUP compares them — fire ratings, gauges, types, board thicknesses — and checks both directions of reference. Every item quotes the sheet and the paragraph, side by side, and drafts the RFI. It reads the written blocks on a sheet: schedules, general notes, keynotes. It does not read geometry.
The scope for each sub, generated from the contract documents instead of last job's copy-paste.
What each sub actually owes — by spec paragraph, by SD classification — issued to them as a request.
The sub opens SPECUP on their phone and the first thing on the screen is what they owe you — with the 7-day resubmittal clock running on anything you sent back. They upload the package and send it. No email chain.
The moment they upload, SPECUP reads the package and tells them which required items it couldn't find — before it ever reaches your desk. It flags for review; it does not rule.
When a package lands you're told what actually arrived: complete, short by a named count, or not verified because nobody could read it. A scan that can't be read is never called complete.
Most of what goes wrong with AI on a jobsite happens before the answer: the documents weren't really read, or they get re-read differently every time you ask. Both are fixed at the front door.
Your specs get broken into their own paragraphs the moment they arrive — 2.1.B stays 2.1.B. Every answer later cites a paragraph you can find in a paper copy, not a floating chunk of a page.
Ask the same question twice and you get the same answer back, marked as the answer of record, with the date it was first given. Told 5/8″ on Tuesday and 1/2″ on Thursday is how a wall gets torn out.
New revision, an ASI, an RFI landing on that section — the recorded answer dies and it's worked again from scratch, and it tells you which of those happened. Consistency that outlives the documents is worse than none.
Old sheets, marked-up prints, anything that went through a copier has no text in it at all. Those get read off the image — and the words stay pinned to where they sit on the page, so you can still see the line.
Schedules, two-column pages, small type on a reduced print — the pages a character reader gets wrong get read a second time a different way, and the two readings are checked against each other line by line.
A line that can't be traced back to the scan is thrown out and counted. An invented schedule row reads perfectly, which is exactly why it can never be allowed to become a requirement.
Open any document and you can see exactly what came out of it — how it was read, how many paragraphs, how many carry a “shall,” and every line a second reader tried to add that didn't match the scan. A half-read spec book should never look the same as a fully read one.
Russell doesn't ask you to take his word for it. Every answer carries the document behind it — the actual page, with the exact line highlighted.
Click through and you get the page as it was uploaded, with the referenced lines boxed. Not a summary. Not a quote he typed. The page.
A 30×42 sheet on a phone is a grey rectangle. So a drawing comes back cropped to the area in question and enlarged, with the whole sheet one tap away.
Scaled off a drawing, interpolated, assumed — it says so, shows how he got there, and drafts the RFI. Scaling is fine. Scaling quietly isn't.
Anything he worked out stays unvalidated until someone checks it and says what they checked. Your name and the time go on the record.
He names the section that should have covered it and whether you've uploaded it. If a document is on the shelf but couldn't be read, that isn't not there — it's we can't tell, and he says that instead.
Ask him a general question and he'll answer — marked as coming from outside your contract, and whether he looked it up or already knew it. The spec still governs.
Every answer is recorded — the question, what backed it, who signed it off. If a reviewer ever asks how you knew, the answer is a document, not a memory.
Every review produces an assembled, review-ready package — put together the way the reviewer expects to receive it. That's the whole point. Load the full plans and specs, or drop in a single submittal with its spec section — same review either way.
The Master Submittal Register generates itself from your specifications — transmittal number, SD classification, paragraph, ball in court.
Every required item gets a cover sheet ahead of its documentation. Missing items get one too, flagged in place at the right register position.
Requirement, paragraph reference, what's wrong. Seven calendar days to resubmit complete. Partial resubmittals aren't accepted.
No fragment swaps. No partial replacement. Every revision produces a full new version, so the record never goes ambiguous.
Your transmittal comes out on ENG 4025 and your register on ENG 4288 — the actual current editions, as real PDFs. Download and attach. Not our format. Theirs.
Nothing we make carries our branding into a deliverable you send. And if your company already has a daily report or a form it likes — upload it, mark the spots to fill, confirm what each field is, and SpecUp fills it from the project record. Your fonts, your letterhead, your logo. We put the values in; we don't rebuild your document.
Status: submittal review, package assembly, the federal transmittal forms (ENG 4025 / ENG 4288) and the plan-against-spec check all run in the app today Now · in-app. Standing up a full project — loading a contract's specs and drawings and building the register with you — still runs as a managed service Now · service.
Born on federal jobs — the hardest compliance in construction. If it holds up to a federal reviewer, your commercial and residential work is the easy day. Turn on what you need; upgrades are a switch, not an implementation.
Bid leveling, inclusions and exclusions, clarification requests, award straight into a live project.
Register, review, three-phase inspections, deficiencies that close the loop.
Talk it in. Punch from a photo. Daily CQC report that assembles itself. No signal required.
Start a walk; everything you capture attaches on its own. Hand each trade their own list at the end.
The contract on one side, what's actually there on the other, dated the day you saw it. We never price it.
Project inbox, triaged. The answer drafted from the record, cited — you approve and it sends.
What each sub owes, what's late, what's outstanding — surfaced before it's a problem.
Deliveries checked against approved submittals. Existence is not compliance.
Field suite on day one.
Turn on the submittal engine when the federal job lands.
Same data. Nothing to migrate.
Published at beta close. Ask and we'll talk straight.
I didn't set out to build software. I set out to stop losing weeks to packages that came back over one missing certificate.
I've worn the hats this tool serves. I've also taught those positions how to use project management software. The information is there and the tools are there — but the flow and the integration were never built for actual field operations. Not for the questions that get asked in the field and need an answer right now. Not for the stop-and-go reporting you get when the day keeps interrupting you.
This got built owning those roles and working in the trenches with QC managers, superintendents and project managers — on federal, commercial and residential work. The design was led by the rigor the U.S. Army Corps of Engineers demands — how work gets reported, filed and logged. What the field needs is consistency, and it needs the answers right now.
I know what a reviewer looks for, what a rejection costs, and how much of a super's week disappears into paperwork that a machine should have handled. So I built the tool I wanted on my own jobs, and I test every piece of it on real work before anyone else sees it.
SPECUP isn't an AI wearing a hard hat. It's the record — what your contract says, and what got decided against it. It knows what a submittal is, what a spec paragraph means, and why a complete package matters. If it doesn't know something, it tells you — that rule doesn't bend.
FounderSpecUp, LLC · Arizona
The rigor came from federal work. The daily work of running a job is the same everywhere.
USACE CQC dailies, three-phase QC, ENG 4025 and ENG 4288 as real PDFs, the register and the released package. Straight answers on RMS, ProjNet, FedRAMP and CUI.
See the federal page →Your super walks and talks; the punch items write themselves and route to the sub. Subs submit their own packages. Your daily report format, filled.
See the commercial page →On the GC's job, what you owe is the first thing on your screen and you see what's missing before you send it. And you don't need a GC to use it. Run your own job — your walks, your first-install inspections, your punch list, your change items.
Two minutes, your own numbers, no email required. If Russell already handles something you're paying people to do, you'll see it on the same screen.
