Flying probe data validation
Golden-data validation across reruns and sites, with BOM alignment and runtime/rework tracking for stronger failure forensics.
Role: Process Engineer / Software Developer — Integrated Test Corporation (ITC), 2023–2025
Stack: Python, test result diffing, BOM validation, operational metrics
// Highlights
- Established and compared golden data across job reruns for drift detection.
- Automated comparison of new runs to historical results and the BOM for multi-site consistency.
- Tracked machine runtime and manual rework to enrich failure narratives.
// Problem
Flying probe results needed consistent baselines across reruns and facilities; without structured comparison, subtle regressions and BOM mismatches were easy to miss.
// Approach
Scripted diffing against golden captures, layered BOM checks, and added lightweight operational logging so engineers could connect electrical findings to line activity.
// Outcome
Stronger confidence in cross-site repeatability and clearer documentation when failures required deeper review.
// AI & orchestration
Cross-site repeatability depends on stable baselines; vendor output drift and field ordering differences can hide regressions.
Pattern: Normalize vendor exports → compare against golden captures → join with BOM expectations → produce human-readable diffs for handoff.
// Technical notes
- Normalization of vendor-specific field ordering before comparison.
- Human-readable diff summaries for handoff to process owners.
// Metrics
Golden-data diffs + BOM alignment for drift detection
Multi-site / rerun comparisons