Vanthrope
Back to Work/Belhide Operational Intelligence
Own Business (Co-owned)Running in production[ev-bh-01]

Belhide Operational Intelligence

Context: Tannery processing facility and direct-to-consumer brand co-owned by the founder, processing high-volume raw hide skins and finished goods.

The Operational Problem

“Over 3,000+ daily raw hide skins were tracked on physical paper tally clipboards, creating a 14% grading discrepancy between incoming intake logs and final inventory ledgers.”

Timeline4 weeks initial build, ongoing production operations
ClassificationOwn business
StatusRunning in production
Truth Gate Evidence[ev-bh-01]
Operational Constraints

Real-world boundaries

▸Harsh tannery floor environment requiring ruggedized tablet input
▸Fast-moving raw skin intake queues
▸Multi-stage wet-blue grading variations
Engineered Solution

What was deployed

Deployed deterministic tablet-first scan workflows with real-time lot yield tracking, replacing clipboards with barcode scanners and automated weight reconciliation.

How it is kept safe
Immediate discrepancy flag if raw lot weight deviates > 2% from bill of lading
Dual-grade supervisor confirmation for high-grade hide batches
Offline-first deterministic sync queue ensuring zero data loss during floor network drops
Measured Outcomes

Verified results

3,000+ daily skins
Method: Real-time lot yield tracking across tannery intake
Evaluation Period: Daily continuous operations
14% discrepancy → 0%
Method: Eliminated grading discrepancies via tablet-first scan workflows
Evaluation Period: Production deployment

“Replacing clipboards with tablet scan workflows eliminated grading disputes at the sorting gate. Every single hide of the 3,000 processed daily is accounted for.”

— Arslan Rehmani, Founder & Systems Developer
What Went Wrong & What Changed

The friction encountered

Friction: Initial WiFi dead zones in damp chemical processing pits delayed scan uploads.

Remediation: Implemented an offline-first IndexedDB buffer that syncs immediately when devices re-enter access points.

Post-Mortem Reflection

What I'd do differently

Would have implemented webhook signature verification and dead-letter queues on day one.

Engineering Stack: Node.js, Next.js, Supabase / PostgreSQL, PWA / Offline-First, Barcode Engine
Next Steps

Ready to diagnose your operational friction?

Schedule a 20-minute operational scoping session directly with our Principal Systems Architect. We evaluate your current transactional volume, assess database readiness, and determine the deterministic architecture required to eliminate your operational bottleneck.

Direct founder response guaranteed within 24 business hours.
Led by Arslan Rehmani, Founder & Principal Systems Architect. For personal engineering field notes, open proof ledger, and interactive diagnostic scans, visit arslanrehmani.com →