Hello All,

Greetings from ZettaMine!!!

Role: ML Data Engineer

Location: Hyderabad (Preferred) / PAN India

Years of Experience: 5–8 Years

Notice Period: Immediate Joiners

Mandatory Skills

- Feature Store and ML training pipeline experience
- Strong Apache Spark, Dataproc, and PySpark
- BigQuery proficiency, including SQL and table design
- Python, Java, or Node.js
- Hands-on PHI / HIPAA data handling
- Specialist ML data engineering experience combining Feature Store engineering, Spark feature pipelines, and HIPAA data handling

Core Responsibilities

- Own the Feature Store population pipeline, ingesting Behavior Signals, Insurance Coverage Signals, Clinical Signals, Engagement Signals, Rx Signals, and Contextual Features from curated BigQuery data marts into the GCP Feature Store.
- Design and maintain ML training dataset pipelines for NBA and AI Insight models, including offline batch paths, online serving feature paths, training/evaluation splits, and dataset versioning.
- Integrate Adobe Analytics event data as a behavioral signal source, aligning it with clinical and benefits data for multi-source model training.
- Operate and tune Dataproc Spark Jobs / Dataproc for large-scale feature engineering and model training data preparation.
- Monitor feature freshness, training data drift, and model data quality in partnership with Vertex AI pipelines.
- Collaborate with backend domain engineers to define Kafka event schemas that feed ML operations.
- Implement HIPAA-compliant handling of PHI across ML training datasets. Apply data classification tagging, field-level masking, and de-identification before model training. Enforce role-based access controls on Feature Store entries and maintain audit trails for PHI access and lineage.
- Support Disaster Recovery (DR) zone design for ML data assets.



Coordinate cross-region replication of training datasets and Feature Store snapshots, define RTO/RPO for ML pipeline recovery, and participate in DR drills.
- Implement data archival policies for ML training datasets. Manage BigQuery dataset expiration schedules, archive historical training snapshots to tiered Cloud Storage, and ensure HIPAA-mandated retention windows are met for PHI-derived features.

Required Qualifications

- 5–7 years of hands-on data or ML data engineering experience in a production GCP environment.
- Solid proficiency in Python, Java, or Node.js for pipeline development, feature engineering scripts, and ML data tooling.
- Strong experience building ML training pipelines and Feature Stores; GCP Feature Store preferred.
- Deep proficiency with Apache Spark / Dataproc / PySpark for large-scale feature engineering.
- Experience with Kafka for consuming streaming event signals into ML pipelines.
- Familiarity with Adobe Analytics data structures and integration patterns.
- Solid BigQuery skills, including complex SQL, window functions, and ML-optimized table design.
- Understanding of the ML lifecycle, feature engineering, data versioning, and train/evaluation splits.
- Experience with Vertex AI Pipelines or similar MLOps tooling.
- Hands-on experience handling PHI under HIPAA, including de-identification techniques, Safe Harbor, Expert Determination, field-level masking in training data, and audit logging for PHI access in ML systems.
- Familiarity with disaster recovery planning for data platforms, cross-region replication, RTO/RPO definition, and recovery testing.
- Experience with data archival and retention for ML datasets, tiered storage policies, and HIPAA retention compliance.

Interested candidates kindly share your updated CV to [email protected] or WhatsApp to (phone hidden).

Thanks & Regards,

Aruna.M

📌 ML Data Engineer (Hyderabad)
🏢 ZettaMine Labs
📍 Hyderabad