HeatCore AI Presentation — Real-Time Thermal Weld Monitoring
Product overview: how HeatCore AI trains on your process, detects anomalies live, and documents ISO 17635-aligned evidence automatically.
Who This Presentation Is For
Welding engineers and quality managers evaluating thermal AI for inline inspection. This video explains how HeatCore AI works—from initial calibration to real-time defect alerts—and why it's built for ISO 17635/3834 workflows.
Topics Covered
- Model training — Teaching the AI what a "good" weld looks like using your process
- Edge deployment — Running inference on the cell, not in the cloud
- Real-time detection — Flagging porosity, lack of fusion, and temperature anomalies
- Evidence logging — Automatic capture of thermal frames, AI verdicts, and timestamps
Why HeatCore AI?
- Catch defects inside the weld cycle, not downstream
- Reduce scrap and rework with real-time operator alerts
- Audit-ready evidence for ISO 17635 and customer quality specs
- Scales from single-cell pilots to multi-plant rollouts
Related Pages
Transcript (short)
Show transcript
This presentation walks through HeatCore AI: mounting and calibrating the thermal camera, collecting baseline data, training an on-edge model on your process, and running live inference to flag anomalies during welding. It also highlights evidence logging (timestamps, frames, verdicts) to support audit-ready workflows aligned with ISO 17635 / ISO 3834.