Manufacturing organizations face rapidly growing demand for visual inspection, process monitoring, and data-driven quality control—yet conventional machine vision systems struggle to scale across thousands of products and global factories. This talk introduces an AI-driven image analytics framework designed specifically for high-variety manufacturing environments, where product diversity, complex assemblies, and global distribution require solutions that extend far beyond traditional pass/fail inspection.

We outline how modern deep learning methods—including segmentation, anomaly detection, classification, and object detection—unlock deeper insight into manufacturing variability and can outperform human or rules-based inspection in challenging cases. Real deployments are highlighted, such as multi-model systems processing more than 250,000 images per day while providing predictive, parameter-based process insight.

At the core of this approach is the Analytics for Manufacturing Imaging (AMI) platform, a connected MLOps ecosystem that integrates data curation, model training, deployment, monitoring, and democratized tooling for engineers. This framework transforms inspection into continuous visual process analytics, enabling faster root-cause understanding, improved quality, and scalable adoption across diverse manufacturing lines.

Speakers

Gary Kunkel Engineering Director, Manufacturing Inspection & Automation Platforms Medtronic (United States)

Gary Kunkel is the Director of Imaging Technology and Analytics within Medtronic’s Enterprise Operations Innovation team. He leads a team at the forefront of developing machine and computer vision technologies for inspecting, characterizing, and monitoring medical devices across all of Medtronic's product’s manufacturing lines.

Before joining Medtronic, Gary was the Director of Recording Head Metrology and Analytics, where his team pioneered cutting-edge metrology and image analytics solutions, achieving ultra-precise measurements at the sub-angstrom level and supporting high-volume production of three million units per day. During his tenure at Seagate, he held engineering and management roles spanning system and sensor design, computational modeling, measurement, test and reliability, and process development.

Gary’s academic background includes a Council on Science and Technology Post-Doctoral Teaching Fellowship in Mechanical and Aerospace Engineering at Princeton University and a PhD in Aerospace Engineering and Mechanics from the University of Minnesota. Despite his passion for aerospace, he has yet to build a rocket that wasn’t made of cardboard.

 

---

Event Details

FORMAT: Oral presentation followed by audience Q&A.
MENU: Coffee, decaf, tea and water will be available nearby.
SETUP: Theater seating.