Principal Engineer - Data Engineering

About the role

Western Digital is looking for a Principal Engineer in Data Engineering in Singapore. The data handled is experimental measurement data from precision product development instruments, not transactional or clickstream data, so the pipelines must meet the needs of scientific data and AI training. The role focuses on building ML-ready scientific data pipelines.

Key responsibilities

• Build and maintain versioned feature engineering pipelines that turn raw engineering, sensor and operational data into structured ML-ready features.
• Design and run data quality checks (completeness, schema consistency, distribution stability, label accuracy) and alert the ML team when quality degrades.
• Maintain training data versioning and lineage tracking, and detect drift between deployment and training data.
• Implement data contracts between upstream producers and downstream ML consumers, including access control and retention.
• Contribute to real-time sensor data ingestion pipelines, developing operational ownership over 6-12 months.
• Build pipeline infrastructure for synthetic data generation (ingestion, storage, versioning).
• Maintain the training dataset registry, feature store and model input validation, integrated with the AI platform (AWS Kubernetes, PortKey, Agent Gateway, LangFuse, AWS Guardrails, Elastic Search).

Required profile

• Bachelor's or Master's degree in AI, Computer Science, Data Engineering, Electrical Engineering, Applied Mathematics or a related field; AI major preferred.
• Fresh to 1 year of experience; demonstrated end-to-end data pipeline projects (academic, personal or internship) are the main evaluation criterion.
• Curiosity and conceptual understanding of how data pipelines feed ML model training; no prior ML work experience required.

Required skills

• Python and SQL, with strong pipeline design fundamentals.
• Batch pipeline architecture, reliability, schema management and fault tolerance.
• Pipeline orchestration (Airflow, Prefect, AWS Glue Jobs).
• Data quality principles, data versioning and lineage, and basic ML data lifecycle awareness.
• Good to have: Feast, Tecton, DBT, DLT, Label Studio, CVAT, Iceberg, Dremio, AWS Lake Formation.