Azure Data Engineer Starlink WLL

• Provision cloud infrastructure as code across multiple environments: event bus, data lake, warehouse

capacity, and supporting services. Cost monitoring baked into every resource.

• Build a deterministic identity resolution service that examines incoming events, resolves identifiers across

source systems, assigns stable person IDs, and runs backward stitching to retroactively link anonymous

events to known persons. Implement SCD2 versioning and hysteresis to prevent identifier flapping.

• Build and maintain a warehouse model with bronze/silver/gold structure covering customer profiles, event

history, orders, products, inventory, consent, segments, and platform operational logs. Tests on every model.

• Implement a consent and suppression layer applied at every audience export — per person, per channel,

per purpose — with a global do-not-contact override.

• Deploy and configure an orchestration platform covering ingestion, identity resolution, transformations, and

activation, with schedules calibrated to source freshness tiers and Slack-integrated failure alerting.

• Build server-side conversion API integrations for major advertising platforms, using platform-specific

identifier hashing and deduplication keys consistent with client-side pixels.

• Implement right-to-erasure at the data layer: both soft delete (PII tombstoned, aggregates preserved) and

hard delete (rows removed). Monthly synthetic test verifies the pathway is functional.

• Configure observability, alert tiers, geo-redundant storage, and document disaster recovery procedures.