Upload a CSV, tell Kirelta which rows were healthy, and get a real verdict from the
real engine — before you write a single line of integration code.
ⓘYour file is read in your browser. Parsing, column detection and the split all happen
locally — nothing is uploaded. Only the numeric rows you select are sent to the API when you press
Run, and they are not stored: Kirelta keeps the fitted baseline, not your data.
Columns
Kirelta reads numbers. Text and ID columns are switched off automatically —
an ID is a label, not a measurement, and feeding it in would make the engine chase a counter.
Columns that look like a score, match, or decision
(final_score, shortlisted, a bare 0/1 flag…)
are switched off too. Those are usually what an earlier model computed, not what came in —
mixing them into the baseline blurs "the resumes changed" with "the accept rate changed," and the
verdict would not tell you which. Turn one back on only if you are certain it is a raw input.
Baseline — rows Kirelta learns as healthy
—
The earliest rows in the file. This is what "normal" will mean.
To check — rows it will judge
—
The rest. Kirelta has never seen these.
Drag to move the split. If your file is ordered by time,
this is exactly the real question: does last month still look like the months before it?
Held in this tab only, never stored.
Get a free key — 2 models, 1,000 checks a day, no card.