Identify abnormal yield movement, repeated failures, and equipment- or process-concentrated quality signals.
Product Quality & Manufacturing Data
I investigate manufacturing quality problems — and turn failure signals into data-driven corrective actions.
Failure analysis · yield movement review · false-fail separation · Python / SQL automation · quality workflow systems
I combine semiconductor product evaluation experience with Python, SQL, statistical analysis, and internal tooling to detect abnormal quality signals, analyze failure patterns, and support corrective-action follow-through in high-volume manufacturing environments.
M.2 SSD DAS FET Burnout Root Cause Analysis
Traced intermittent DAS FET burnout in M.2 SSD LI testing to conductive debris, PCB/socket alignment risk, and DAS screening limitations, then helped reduce collected/confirmed burnout from 11 ppm to 2 ppm.
Demonstrates: root-cause chain analysis, screening logic improvement, statistically validated defect reduction
Manufacturing Quality Anomaly Detection Dashboard
Synthetic dashboard for yield movement, defect-code concentration, equipment-level failure patterns, false-fail signals, anomaly ranking, and corrective-action validation.
Demonstrates: statistical signal detection, ML-assisted prioritization, engineering review workflow
UFS Yield Excursion Analysis
Multi-month NGBin 19/109 trend review connecting backend MBT failures to wafer-edge process patterns.
Demonstrates: yield excursion review, process-pattern investigation, wafer-edge trend analysis
SMT Calibration Quality Case
From LI 4523 symptoms and MLCC damage to an upstream SMT process-control signal.
Demonstrates: downstream symptom tracing, upstream process-control signal detection, corrective-action monitoring
Manufacturing Quality Tableau Dashboard
Synthetic-data Tableau dashboard for yield trend, defect Pareto, equipment concentration, retest behavior, and corrective-action validation.
Demonstrates: dashboard-based quality review, abnormal-signal communication, post-action validation
Method
How I work
Use test logs, retest behavior, failure-code distribution, and process history to separate false failures from true product-quality risks.
Translate data evidence into issue ownership, corrective-action proposals, and cross-functional review points.
Monitor post-action yield movement, defect recurrence, and quality-risk indicators to confirm whether the action was effective.
Systematize repeated reviews into scripts, dashboards, and workflow systems so quality signals remain visible over time.
Target Role Fit
Product Quality / Failure Analysis / Manufacturing Data / Corrective Action
This portfolio is most relevant to roles involving:
- Product quality improvement in high-volume manufacturing
- Failure analysis support and defect-pattern investigation
- Yield movement review and false-fail separation
- Manufacturing data analysis with Python / SQL
- Quality monitoring dashboards and automated reporting
- Corrective-action tracking and post-action validation
- Internal workflow systems for product evaluation and production-readiness review
Problems I enjoy solving
- When first-pass failures do not tell the whole quality story
- How to separate false failures from true product-quality risks
- How to trace downstream test symptoms to upstream process drift
- How to detect equipment- or process-concentrated defect patterns
- How to make hidden calibration variation visible before defects recur
- How to turn repeated quality reviews into automated monitoring workflows
Let's Work Together
Open to roles where manufacturing data, failure analysis, and software tooling are used to improve product quality and engineering decision-making.