Key Takeaways
- James Bayly says Mindhive’s AI grading engine—trained on 14 million hides—detects 35+ defect types with high precision.
- Italy and Brazil were prioritized for expansion based on finishing sophistication and large-scale, multi-site operations.
- Series A capital will fund team growth, R&D, and regional support infrastructure in Brazil and Europe.
- Early pilots of FinishSelect™ show significant yield gains and faster processing speeds for finishing operations.
- Adoption barriers are organizational rather than technical, requiring deep change management.
James Bayly on AI Transforming Traditional Leather Inspection
Mindhive Global CEO James Bayly characterizes the company’s technology as a substantial shift from rapid human judgment to digital, data-rich analysis. The platform uses structured lighting, 4K line-scan cameras and neural networks to capture and grade each hide objectively.
“Our core technology utilizes lights, cameras and significant computing power to enable machine-vision technology to capture an image of each hide we grade,” Bayly explained. Mindhive’s BlueSelect™ system analyzes these images “through neural networks trained on 14 million hides over 14 years,” giving it the ability to detect “35+ different defect types, from sub-millimeter tick damage to large-scale defects like neck hump or major staining.”
He contrasted this with manual grading, which offers only a brief visual sweep. “A grader needs to assess an area larger than a pool table in less time than it takes to read this sentence… The traditional method does not allow any digital or objective record.” By comparison, Mindhive creates “a complete digital record for every hide that can be reviewed, analyzed, and used to defend grading decisions objectively.”
