Mortgage QC · AVAILABLE FOR EVALUATION
Mortgage Quality Control, on the Mortgage Intelligence Platform.
Review the whole file, preserve the evidence, and focus human judgment where it matters. Running on real files today, on sample, synthetic or de-identified historical files, with every finding carrying the rule version and the source page behind it.
- Source
- Fact/Context
- Policy
- Calculation
- Decision
- Evidence
Why QC is hard
Most defects live between the documents
Experienced reviewers already do this work, and do it well. The constraint is not skill: the checks are numerous, the file is long, the policy keeps moving, and the time is finite.
Qualifying income does not reconcile across sources
- Finding severity
- Significantassigned by the rule at run time
- Review state
- Raised → Acceptedreviewer disposition, 2026-09-08
- Determination
- Not a defectfinal, after review
| Source | Page | Value | Basis |
|---|---|---|---|
| Paystub, 03/2026 | p. 1 of 2 | $8,940 / mo | derived, year-to-date annualized |
| W-2, 2025 | p. 1 of 1 | $8,875 / mo | derived, annual wages / 12 |
| URLA 1003 | p. 2 of 9 | $9,200 / mo | stated by the borrower |
| Transmittal 1008 | p. 1 of 2 | $9,200 / mo | used to qualify |
- Basis for the determination
- Within tolerance, and the reviewer accepted the basis. Raised because the 1008 used the stated figure rather than a derived one, and which basis stands is a reviewer’s call, not the rule’s.
- Evidence kept
- Rule XDC-INCOME-RECONCILE v2.3.0, published 2026-07-14, active at run time. Four values, each with document, page and position. Program rule set and MISMO field path recorded. Rules that could not run listed separately.
- Reviewer disposition
- Accepted · QC reviewer · 2026-09-08
“Underwriter used the stated figure with a written calculation on file.”
Severity states what the finding would mean if it stood. The review state records what the reviewer did with it. The determination is the final answer. A reconstruction of the working application on a synthetic loan; the values, sources, threshold, states and disposition are the ones the product records, and the rule’s executable logic is not shown.
No single document is wrong. The finding only exists when the file is read as one thing. That is what cross-document reasoning means, and it is where the platform shows first.
- What a miss costs
- $32,288Estimated average cost per repurchase demand. National Mortgage News, drawing on STRATMOR Group analysis.
- What the rate hides
- 1.50%Critical defect rate, CY2025, flat against 1.52%, while eligibility defects rose 291% and credit 166%. ACES Mortgage QC Industry Trends.
- What a loan costs to produce
- $11,109Fully loaded total production cost per loan, not QC cost. MBA Quarterly Performance Report, Q3 2025. No per-loan QC figure is published industry-wide.
- What changes
- Designed for every loan, not a sampleFull-file, cross-document checks rather than a sampled read, so depth stops being the variable that quietly gives. What is on offer today is a controlled evaluation: see the capability matrix below.
Whole-file workflow
Seven stages, and two boundaries
Issues found across the whole file, checked against the applicable requirements, resolved by a person, and every step kept with the finding.
Intake
Loan file received and queued.
Classification
Documents identified and segmented.
Extraction
Values located, kept with page and source.
- Verify: a person corrects what the system is unsure of, before rules run
Synthesis
Reconciled across the file; conflicts recorded.
Validation
Versioned rules run. Skips recorded too.
Review
Exception raised. A reviewer resolves it.
- Findings: what the reviewer opens
- Disposition: a named person decides, and it is recorded
Evidence
Finding kept with the rule version that made it.
- Report: the exported record, with the rules that did and did not run
Every stage, followed through one synthetic file with four findings, is on the walkthrough; what is kept with a finding is taken apart on anatomy of a finding.
Controls
The policy the rules enforce, encoded and versioned
Not a checklist held in someone's head. Each control is a stored rule, bound to a loan program, and the record shows which ran and which were skipped.
Rule sets are selected by loan program, so an FHA file and a bank-statement file do not run the same checks. Which rules were skipped, and why, is recorded alongside the ones that ran. Encoding a published requirement is not endorsement by, affiliation with, or certification from any agency or GSE.
Findings
Four outcomes are recorded, not two
Every rule returns one of four states, and the distinction between the last three is the point.
A system that reports only pass and fail turns data required and not applicable into silence, and silence reads as clean. Each finding carries a severity, the rule and its version, the values used and their sources, and a status a reviewer moves it through.
Evidence
What is kept with every finding
Six months after a file is cleared, someone asks why. A finding that cannot answer is an opinion with a timestamp.
Recorded on every run: the values used, with the source document each came from; confidence per value; which source won where documents disagreed, and on what basis; the MISMO field path; the rules that ran; the rules that were skipped; the loan program applied; and the rule set version. Anatomy of a finding, taken apart in full.
What is retained today: findings and synthesized values keep their lineage to the source data and the source document. What is still being built: presenting that lineage down to page and field level inside the reviewer interface. The record exists; the interface for reading it at that granularity is on the roadmap.
Governance
“You are using AI. Show us what it did, and how you knew it was right.”
Fannie Mae's LL-2026-04 took effect 6 August 2026; Freddie Mac's Bulletin 2025-16 has been live since 3 March. Both require a documented AI/ML governance program, and both hold you to the same standard for the AI your vendors use.
What is the AI doing, and why
Each stage is named and bounded. The AI helps read the file and identify issues across it; it does not make the loan decision. Findings are tied to the applicable requirement, and a named person decides. The purpose of the AI is answerable in one sentence rather than described.
What safeguards are in place
Every finding is tied to the requirement applied and its version, every value to its source page, and uncertain values are surfaced for a person to correct before findings are produced. That is a real safeguard and a bounded one. It does not make a reading error harmless.
How do you know it got it right
Every value carries the source document it was read from, and every finding the numbered rule version. Given the same inputs and version, the same result follows, so a disputed answer is traceable to a specific value on a specific page.
Can you evidence it for a vendor
The obligation extends to vendor and subcontractor AI, held to the same standard. The record exists as a by-product of the review, not as a document written afterwards.
The platform produces the evidence; the lender holds the obligation. Nothing here implies approval or endorsement by Fannie Mae or Freddie Mac.
Current capability
What runs today, and what is still being built
Stated as two columns rather than one, because the distance between them is the thing a buyer needs and the thing vendors usually hide. Current as of the September 2026 capability review.
Running today
These run in the deployed environment now. A rules specialist or an audit lead can be shown them without qualification.
The strongest part: rules, versioning and lineage
The rule model, execution, explainability and lineage can be demonstrated now; the authoring and approval interfaces around them are in the group below.
Implemented, awaiting verification
Everything in this group is implemented; what it has not yet had is demonstrated behavior we would put in front of you as evidence.
In development
Where current and target are furthest apart, stated so timing can be planned rather than discovered.
Intake
How a loan file reaches the platform, today and next
Split into what runs now and what is on the roadmap, because the difference decides what an evaluation can actually cover.
Evaluation and data
How to test it against controlled files
A synthetic file with no data from you, or three to five of your own de-identified files whose outcomes you know. Misses, false alerts and time, agreed in advance and computed together. Your documents are not training data; they are processed inside our cloud environment; retention is contractual.
See where it stands, on a real file.
A working session on a synthetic file, or de-identified files of your own. You will see the parts described above as running, and the parts that are not.