Lanes, equipment, accessorials, pickup/delivery windows — all hand-keyed into the TMS by the dispatcher. ~2 minutes apiece.
The AI follows the same logic a human would: spots the header, reads the body, sums or validates as needed. No template setup. No re-training when the format changes.
{
"broker": "Bridgeway Logistics",
"carrier": "PNW Freight Inc.",
"mc_number": "442918",
"load_number": "BR-2026-7714",
"pickup": {
"location": "Tacoma, WA 98421",
"date": "2026-05-22",
"window": "06:00 — 11:00"
},
"delivery": {
"location": "Salt Lake City, UT 84104",
"date": "2026-05-23",
"window": "12:00 — 16:00"
},
"equipment": "53' dry van",
"commodity": "Packaged goods, palletized",
"weight_lbs": 38400,
"rate": {
"linehaul": 2150.00,
"fuel_surcharge": 315.00,
"total": 2465.00
}
}Once docs are structured, finding anything is a sentence away. Ask in plain English. No SQL, no filters, no Excel pivot tables.