Computer vision-powered real-time tracking for manufacturing logistics
A global HVAC controls manufacturer faced a critical supply chain blind spot:
Designed and deployed an end-to-end computer vision system that automatically captures shipment timestamps at the exact moment pallets exit facility doors. The system combines edge computing, AI-powered label detection, and cloud integration to provide real-time shipment visibility.
Time-of-flight sensors detect pallet entry (4-6m)
4K camera captures frames with shipping labels
Computer vision identifies and localizes labels
OCR decodes barcodes and extracts tracking IDs
Green light confirms successful scan
Data synced to cloud and ERP system
Selected and evaluated cameras, compute units, and sensors. Calculated resolution requirements for reliable barcode detection. Managed vendor partnerships achieving 50% cost reduction.
Led installations at multiple US manufacturing facilities. Coordinated with facility teams on mounting, networking, and environmental requirements. Ensured standards compliance.
Designed edge-to-cloud pipeline architecture. Integrated with client ERP systems for metadata matching. Built APIs and webhooks for real-time data flow.
Optimized processing pipeline with parallel workers. Implemented memory management and failsafes. Migrated from WiFi to Ethernet for stability. Reduced logging overhead.
Established monitoring with Scalyr. Created health check systems and alerting. Documented SOPs for operations team.
Primary technical contact for client stakeholders. Coordinated with AI engineers, platform team, and project management. Trained warehouse associates on new workflows.