Requirements
* Experience designing and operating production data pipelines at scale
* Strong SQL and Python skills and hands‑on Airflow experience
* Experience with both batch and streaming data architectures
* Familiarity with cloud data platforms and distributed processing systems
* Experience modeling event‑driven data for analytics and experimentation
* Knowledge of ad tech data concepts such as auctions, bids, pacing, and yield
* Strong fundamentals in data quality, lineage, monitoring, and governance
* Ability to partner effectively with cross‑functional technical and business teams
What the job involves
* In this role, you will build and maintain data pipelines and analytics infrastructure that power Roku’s CTV advertising auction platform
* You will support marketplace optimization by delivering high‑quality datasets for advertiser performance, publisher yield, and revenue analysis across batch and near real‑time workflows
* You will partner with product, analytics, and data science teams to translate business needs into well‑modeled, reliable data products
* You will contribute to experimentation, simulation, and reporting foundations that inform bidding strategy and monetization decisions
* This role is ideal for an engineer who enjoys high‑scale data systems, strong data quality practices, and measurable business impact
* Design and maintain data pipelines and analytics infrastructure supporting Roku’s CTV advertising auction platform
* Build ETL/ELT workflows in Airflow to process auction events including delivery, bids, impressions, pricing, budget usage, and frequency cap signals
* Create scalable batch and streaming pipelines for billions of daily ad events with strong freshness, accuracy, and schema consistency
* Model datasets for multi‑objective optimization and marketplace analytics across advertiser, publisher, and platform outcomes
* Design aggregated tables and materialized views to support closed‑loop analysis of auction results and bidding behavior
* Own integrations from DSPs, programmatic exchanges, and direct campaigns and standardize schemas for unified reporting
* Partner with data scientists, analysts, and product teams to deliver clean, discoverable, and trusted datasets
* Enable yield and gross profit analysis through dimensional models, win‑rate metrics, demand health indicators, and experiment measurement tables
* Build data foundations for A/B testing analysis, auction simulation, offline replay, and post‑campaign reporting
* Implement data quality checks, monitoring, and observability while supporting privacy and governance requirements
#J-18808-Ljbffr