An officer reads your file
in eleven minutes.
Officer Simulator streams a criterion-by-criterion review in an officer's voice, then a separate final-merits pass — so you see the read before you file, not after.
Inside the simulation.
Choose the officer reading your file.
Neutral, skeptical, seasoned, and judge officers don't read the same file the same way — pick one and the commentary genuinely restyles to that voice, not just a label swap. A skeptical read on an exhibit flags exactly what a neutral read would let pass. Run the same criterion through more than one persona to see where the read actually diverges.
Every exhibit, opened and checked off.
The reading desk fills with your evidence in the order an officer actually works through it, and each exhibit flips from "Waiting" to "Read" the moment the commentary closes its segment — not before. You watch the file get worked through, not just a final list of what was included.
A posture chip for every criterion, live.
Each criterion carries its own posture chip — dashed and ungraded until the commentary on it resolves, then it flips to likely met or needs more evidence, in the open, as it happens. You watch the read happen, not just its final output. Every line also names the exact regulation it's weighing the claim against, like 8 CFR 204.5(h)(3)(v).
Compare your file's read across two versions.
Re-run the simulation after you add evidence and compare it against your last session — which criteria moved, and which one changed your posture. It's the same comparison an officer implicitly makes if your file gets a second look, except you get to see it before they do.
You don't find out how your file reads until an officer already has.
You can't watch a cold read.
You can't watch a real officer read your file cold — only guess at what they'll notice first.
Strong criteria can still read weak.
A general review can tell you a criterion looks strong. It can't tell you the file reads weak in the aggregate anyway.
By the RFE, it's too late to rehearse.
The read that produces an RFE already happened by the time it arrives — and practicing on your own case with your own eyes coaches you on details you already know, not the register a stranger reacts to.
Why generic AI tools fail here.
A guess and a graded, evidence-linked read are not the same kind of answer.
A chatbot doesn't read like an officer.
General-purpose AI answers the question you ask. It doesn't hold an adversarial, criterion-by-criterion posture the way an officer reading your file actually does.
A single pass hides the real risk.
Meeting every criterion individually and reading weak in the aggregate are two different failure modes. Most review tools only ever check the first one.
Feedback without a decision preview stays abstract.
Knowing a criterion is "strong" doesn't tell you what an approval, an RFE, or a denial actually sounds like for your specific file.
Merito shows you the read before the officer does.
How a session runs.
Choose who reads your file.
Neutral, Skeptical, Seasoned or Judge: one reader per session, reading the structured Case File built from your documents, not raw uploads.
The reading desk fills.
The exhibits fill a strip, and each is marked read when the note on its criterion is done. You see the file get worked through, not just a final list of what was included.
Commentary streams, with posture attached.
Criterion by criterion, each note ends in a stamp: reads as met, could use more, or not shown yet, with a quote from the exhibit it rests on.
Preview the decision.
Approval reasoning, a simulated RFE and a denial discussion, each generated when you ask for it, always watermarked as simulated and educational, and drawn from the same Case File.
SIMULATION — EDUCATIONAL DRAFT
Eleven minutes,
criterion by criterion.
Commentary streams in an officer's voice as your file loads onto the desk, criterion by criterion, then a separate final-merits pass closes the session. See the read before you file, not after.
See the read before you file.
Start your free trial