AI-generated scrap procurement quote-and-quality ledger

AI-generated scrap procurement quote-and-quality ledger: per-lot offers carrying ISRI-style grade (HMS1&2 80:20, Shredded Scrap #1, H2), chemistry constraint (Cu < 0.15%), unit price with Incoterms and port ($378/MT CFR Mundra, $395/MT FOB LA, $402/MT CIF Chennai), verification attributes (ISO 9001, ISRI, bulk-ready) and a composite supplier reliability score (Breven Score, 91-97 on the visible examples) built from reliability, COA history and pricing. Around it the platform generates vision-labeled scrap-lot images (defect and tramp-metal/copper contamination detection pre-shipment), predicted chemistry with risk probabilities, parsed PDF/email RFQ and quote pairs, AI voice-call negotiation records, and cost-per-liquid-ton melt outcomes.

Open buyer room →

Sample

Brickroad customers

Sample this dataset before you buy it.

Your sourcing agent asks Breven and files the sample in your catalog — private to you.

Start 7-day trialNothing is charged until the trial ends.

Coverage

Universe, instruments and categories.

Industry: SteelApplication Software
Instruments: commoditiesequities
Categories

Dataset card

Type: not stated
Format: Multimodal — tabular quote/PO records, vision image data with contamination and defect labels at yard edge, parsed documents (PDF quotes, email chains), and audio (AI voice negotiation calls)
Volume: Currently small but unusually dense: a stated supplier universe of 400+ scrap dealers with grade/chemistry/price/delivery attributes, quotes with incoterm and port per lot, plus vision captures per inspected load and per-lot COA history. Volume scales with the yard-imagery stream rather than quote count, and that stream is the interesting quantity. Request further research on how many loads have actually been vision-inspected to date.
Users: Very early — no public user count. A 1-employee company record plus $1M+ customer-ROI claims rather than installed-base claims suggests single-digit to low-teens mill deployments. Request further research on paying mill count.
History: 2 years
Update frequency: not stated
Growth: Active, pre-scale — 2026 conference circuit presence (CSCMP Edge Europe featured speaker, AISTech 2026 technical contributor, AIST Pittsburgh and Midwest chapter engagement) indicates active selling into steel, while headcount and funding disclosures remain minimal.
Launched: not stated
Delivery: not stated
Entity mapping: not stated
Sample: not stated
Point-in-time: not stated
Licence: not stated

supplier index read from their site