ML & ETL Engineer — Yang

Building AI-driven quality control and data pipelines for smart factory operations

Organization Yang
Role Machine Learning & ETL Engineer
Domain Smart Manufacturing, IPQC Systems
Stack Computer Vision, ETL Pipelines, Tableau, Production ML

Overview

Delivered AI and ETL systems for smart-factory operations at Yang, including a 24/7 in-process quality control (IPQC) audit system and enterprise-scale data pipelines. The solutions enabled real-time defect detection, automated reporting, and data-driven decision-making across more than 20 vendor manufacturing sites.

Key Responsibilities

AI-Driven Quality Control: Led the design and deployment of computer vision-based IPQC systems achieving 95 percent defect detection accuracy, operating continuously to monitor manufacturing quality in real time.
Enterprise ETL Pipelines: Architected robust ETL pipelines processing over one million manufacturing data points daily, integrating sensor telemetry, quality metrics, and operational parameters from more than 20 vendor facilities.
Business Intelligence Dashboards: Built comprehensive Tableau dashboards with real-time KPIs, trend analysis, and predictive insights deployed across 20 plus manufacturing sites for executive and operations leadership.
Automated Reporting: Established automated reporting infrastructure that eliminated over 200 hours per week of manual inspection, data collection, and analysis.
Pipeline Optimization: Optimized data workflows and processing pipelines, achieving approximately 80 percent faster processing speeds and enabling near real-time insights across manufacturing operations.
Cross-Functional Integration: Worked closely with operations, quality control, and IT teams to integrate AI systems into existing manufacturing workflows, modernizing traditional quality control processes without disrupting production.

Technical Achievements

24/7 IPQC System: Delivered a continuously operating IPQC audit system saving approximately 200 hours per week in manual inspection effort.
Multi-Site Deployment: Deployed dashboards and reporting systems across more than 20 vendor manufacturing sites.
High-Volume Data Processing: Implemented automated ETL pipelines processing over one million data points per day.
Model Accuracy: Achieved 95 percent defect detection accuracy using AI-powered inspection models.
Scalable Architecture: Established a scalable data and ML architecture supporting multiple manufacturing facilities and vendor integrations.

Impact

Transformed manufacturing quality control processes through AI-driven automation and data engineering. The platform enabled Yang to maintain consistent quality standards while significantly reducing manual labor, improving visibility, and scaling operational efficiency across its supplier ecosystem.

Outcome

This work demonstrated how tightly integrated AI, ETL, and analytics systems can modernize traditional manufacturing environments, turning fragmented factory data into actionable, real-time intelligence at enterprise scale.