Supplier-portal scraping for compliance documents, ASN updates, and invoice status across hundreds of vendors — feeding a single supplier-intelligence view.
Reach data nobody else can.
Half the data the enterprise actually needs sits behind a login on someone else’s portal — supplier intelligence, customs schedules, regulatory filings, exchange feeds. Quarry is what we built to get at it: headless extraction, antibot resilience, normalisation, and an audit trail on every record. It’s the layer that makes Smelter’s pipelines possible when the upstream system doesn’t do APIs.
What we deliver
- Headless extraction of supplier-portal data
- Regulatory and government-database ingestion
- Customs and trade-feed monitoring
- Financial-feed normalisation (banks, custodians)
- Antibot resilience, captcha handling, rate-limit awareness
- Lineage and audit trail on every extraction
Tools and infrastructure we ship on.
- Headless browsers
- Proxy infrastructure
- Smelter (downstream pipelines)
- OODA (workflow trigger)
From source map to a stable feed — in weeks, not quarters.
Quarry runs in production at portals that change layouts overnight. The work isn't the first scrape — it's the second hundred. Our process is built around that.
- 01
Source mapping
Walk the target systems with the team that uses them today. Document login surfaces, navigation paths, exception screens, and the data shapes that need to come out the other end.
- 02
Extraction design
Build the headless flows behind a stable interface — captcha handling, proxy rotation, rate-limit awareness, and human-fallback paths for the cases that can't be automated cleanly.
- 03
Normalisation + lineage
Every record lands with provenance: which source, which session, which page render, which timestamp. Downstream pipelines (Smelter) receive a typed, validated payload.
- 04
Operate + maintain
Layout changes get caught by health checks before users notice. Managed-services teams adapt extractors, monitor SLAs, and ship fixes inside the change window.
Where Quarry tends to earn its keep.
Highest-leverage deployments by industry — where the source has no API and the data underpins a regulated process.
Regulatory and exchange feeds where APIs lag the portal — corporate actions, KYC/CIN registries, sanctions lists — pulled with audit-grade lineage.
Customs duty schedules, tariff updates, and country-of-origin rule changes monitored across jurisdictions — surfaced to the tax and logistics teams in hours.
Bring us your hardest build.
30-minute discovery call. No decks, no boilerplate — senior engineers in the room, the same ones who'd own delivery.