Start with scan fundamentals and controllability/observability principles.
Learn compression/ATPG before tuning coverage targets.
Study timing, power, and physical integration before release signoff.
Use interview pages to practice mechanism-first communication.
METRIC -> MECHANISM -> EVIDENCE -> OWNER -> FIX -> REGRESSION
1. Which context fails (mode, lot/corner, tester program)?
2. What exact metric regressed?
3. What mechanism explains this metric?
4. Which artifact proves/disproves the mechanism?
5. Who owns the next change?
6. What regression matrix must pass before release?