Software Quality Engineer 軟體品管工程師_TC30461

1. ETL Pipeline Engineering and Operations (40%)

• Maintain and scale daily ETL/ELT pipelines using workflow or scheduling tools.
• Build extractors for manufacturing reports in multiple formats (Excel, CSV, email attachments, compressed archives).
• Implement incremental loads with idempotent transforms (deduplication, checkpoints, rerun-safe jobs).
• Optimize PostgreSQL batch ingestion (COPY / UPSERT) for large volumes with stable memory use.
• Monitor scheduled production runs; handle failures, alerting, and recovery.

2. Data Services, Analytics, and Reporting (25%)

• Maintain backend data APIs and services that power analytics dashboards and reports.
• Write and tune PostgreSQL queries (CTEs, window functions) for operational and analytical use cases.
• Support dashboard / reporting needs: KPI definition, trend analysis, and ad-hoc data requests.
• Standardize manufacturing metrics such as station yield, First Pass Yield (FPY), and repair turnaround time (TAT).
• Apply basic statistics or data science methods (e.g. anomaly detection, SPC-style quality tracking) where they improve insight.

3. Data Governance and Cross-site Communication (15%)

• Serve as the technical contact for data exchange with contract manufacturing sites.
• Align report formats, column schemas, and naming standards with site engineering teams.
• Improve data quality on incomplete or inconsistent fields (e.g. defect classification, part numbers).
• Manage master data versioning (e.g. BOM) and coordinate automated report collection.
• Investigate source-data issues: template changes, corrupted files, and encoding / formatting errors.

4. Database Administration and Warehousing (20%)

• Design and maintain warehouse layers: staging, ODS, DW, and application summary tables.
• Manage PostgreSQL schema changes, indexing, and routine database maintenance.
• Enforce role-based access control for ETL, BI, and administrative users.
• Maintain data quality checks, pipeline tests, and documentation.
• Own historical backfill procedures: backup, reset, reprocess, and verify.