How it works

An inquiry in, a defensible answer out — with receipts.

These are real screens from the product, not illustrations. Each is captioned with exactly what the machine did.

1 · Parse — the customer's words become typed constraints

GRADE reads the inquiry and extracts a structured list of hard constraints: a particle-size limit as a number, a chemistry as a string, PFAS-free as a boolean. Every extracted value carries a verbatim quote from the inquiry. If the quote is not present in the text, the value is rejected — the model cannot invent a requirement.

2 · Screen — deterministic SQL, never a guess

The constraints become a database query. A maximum particle size is a WHERE clause; it is never compared by embedding similarity. A product that fails a stated constraint cannot survive. A product missing data on a constrained field is flagged unknown for a rep to check — never silently passed.

GRADE screening view showing PASS and FAIL verdicts with reasons and source quotes
What the machine did: screened 48 products; two survive; each eliminated grade shows the exact reason — "Mean particle size 25 µm below the 40 µm minimum" — and each constraint shows the customer's own words beneath it.

3 · Rank — the model only sees survivors

Only products that passed the deterministic screen are handed to the model for ranking and rationale. It orders them and explains the fit; it cannot resurrect a grade that failed a hard constraint. Every recommendation still traces back to the screen.

GRADE run report with ranked recommendation, rationale, and full screening detail
What the machine did: ranked the two survivors, wrote a rationale and a risk note for each, and produced a printable report a rep can attach — with the full elimination list, every reason intact.

4 · Draft — a reply your rep approves

GRADE writes a reply the rep reviews, edits, and sends. Nothing sends automatically. When the rep records what the customer actually bought, that outcome feeds a measured-accuracy number — the honest version of "it learns," scoped to that customer's own data.

GRADE home dashboard with KPIs, recent runs, and system status
What the machine did: logged the run to the queue, tracked it from drafted to sent to resolved, and rolled measured usage — turnaround, approvals, provenance coverage — onto the rep's home screen.
The screen is the trust layer, so it is deterministic and identical for every customer. There is no adjustable strictness, no "prefer this product even if it fails a constraint." Read exactly how we hold that line on the How we test page.