The Marketing Data Engineer will build and maintain ETL pipelines and data-quality controls for advertising platforms using SQL, Python, and cloud data warehouses. This role ensures the reliability of marketing dashboards and reporting by reconciling data across various ad-tech sources.
Our client in the US is looking for a Marketing Data Engineer to own the data pipelines and data-quality foundation that power their dashboards, pacing outputs and recurring reporting.
Core responsibilities
Build and maintain API/ETL ingestion from DSPs, ad servers and other advertising platforms.
Normalize campaign data and maintain metric definitions used across dashboards and reporting.
Manage cloud-warehouse structures and client data deliveries where required.
Reconcile pipeline output against raw platform exports and investigate material variance.
Build automated data-quality checks, alerting and monitoring for pipeline/dashboard health.
Manage service-account/API credential workflows in the clients owned environments.
Support platform migrations and rebuild data integrations without reporting discontinuity.
Partner with Dashboard Developer and Programmatic Lead on definitions, mapping and release validation.
Requirements
Must-have profile
4+ years in data engineering/analytics engineering with production ETL/API responsibility.
Strong SQL plus at least one production programming/scripting language such as Python.
Experience with cloud data warehouses such as BigQuery, Snowflake or equivalent.
Experience reconciling data across multiple source systems and designing data-quality controls.
Ability to own production pipelines, troubleshoot failures and document data definitions clearly.
Preferred experience
Direct experience with advertising/marketing platform APIs and campaign data.
Experience with DV360 or other DSP data models, ad-server data or marketing attribution feeds.
Experience with scheduled reporting/alerting and secure client data delivery.
What success looks like
Reporting data is reliable and reconciles within agreed tolerance.
Pipeline failures or data lag are detected before clients notice.
Platform changes do not create reporting gaps.
Metric definitions remain consistent across dashboards and recurring reports.