Customs & trade-document extraction
BOL, commercial invoice, packing list, and customs-declaration extraction with structured output and exception routing — tuned to the UAE / GCC document set, not a generic OCR wrapper.
We run B2B marketing and AI automation for UAE logistics and trade operators — freight forwarders, 3PLs, and platforms moving cargo through Jebel Ali, JAFZA, and DAFZA. Shipper-intent demand on one side; customs-document extraction, exception routing, and quoting copilots on the other.
The UAE is a re-export hub — cargo moves fast or it costs money sitting in a yard. The operators who win clear documents faster, quote faster, and catch the exception before the demurrage clock starts. That’s an automation problem before it’s a headcount problem.
We build BOL, commercial-invoice, and customs-declaration extraction with structured output, exception-routing agents that surface the problem shipment before the SLA breach, and quoting copilots that draft against your rate card and capacity — reviewed by a human, never autonomous on a commercial commitment.
Finyki is remote-first in GST hours, fluent in the UAE trade stack — Mirsal / Dubai Customs flows, free-zone processes, and the EDI dialects carriers actually send. Priced in AED.
Every engagement targets a shipper search pattern and an operational exception — we ship both.
BOL, commercial invoice, packing list, and customs-declaration extraction with structured output and exception routing — tuned to the UAE / GCC document set, not a generic OCR wrapper.
Incoming event → classification → suggested action → human review → resolution, logged. Replaces the daily standup that surfaced the problem shipment 18 hours late.
Agent-assisted quoting that drafts against your rate card, margin floor, and capacity. Edited by a human; never an autonomous commercial commitment.
Search programs for lane, mode, and free-zone-specific shipper intent, AEO-tuned for the answer engines enterprise shippers now use to source 3PLs.
LinkedIn, search, and targeted content reaching supply-chain leaders and shipper procurement across the GCC. Scoped against pipeline, not raw MQLs.
A dashboard connecting quote to booking to margin — win-rate by lane, capacity utilisation, and a forecast operations and finance both trust.
Demo on clean data. Production trade documents are filthy and inconsistent. The agent that hit 95% in the pilot hits 60% live unless it was built and eval’d on your actual document set. We build on the real thing.
Nobody integrated to the TMS or customs flow. An extraction agent that emails ops isn’t automation. Integration to the TMS, the customs flow, and the CRM is the whole job — we don’t skip it.
EDI treated as a conversion target. EDIFACT and the carrier dialects don’t disappear because you put an LLM in front of them. We treat EDI as a first-class input and build parsers that survive the inconsistencies.
ROI pitched as FTE cuts. Real payback is clearance cycle-time and demurrage avoided, plus booking-rate lift on shipper search — not headcount theatre. We scope for the former.
Replaced an overnight manual triage workflow with document extraction and an exception-routing agent that surfaces the problem shipment before demurrage starts — humans resolve, the agent watches.
Built the RFQ-response copilot drafting every tender against the rate card, plus the lane-intent SEO and demand program targeting enterprise shippers. Human reviewer per quote; zero autonomous sends.
45 minutes, free. We find the right first lever — no slides.
One page: the metric, the plan, and the cost in AED.
Working output by week two, eval’d against real data.
Documented and owned by your team. Retainer optional.
Yes. We’ve built extraction and validation for customs declarations, commercial invoices, and packing lists tuned to the UAE / GCC document set and free-zone processes. Accuracy is gated on evals against your real documents.
Yes — integration is more than half the work on a logistics engagement. We wire the TMS, the customs flow, and the CRM, document it, and hand it over.
No. Anything with a commercial commitment — quotes, SLA-bearing ETAs — gets a human reviewer. Informational updates can be autonomous with monitoring.
First production system typically AED 100k–280k; retainers from AED 25k–65k / month. The integration surface and document variety are the biggest variables. Excludes 5% VAT.
Document and exception automation usually show measurable cycle-time and demurrage impact inside 6–8 weeks. We scope against the metric your COO is graded on.
Forty-five minutes, your agenda, no slides. You’ll leave with a clear read on the right first move for your Logistics & Trade team — and a rough shape for the engagement if there is one.