• Design, build, and deploy machine learning solutions for audience targeting, lookalike generation, and individual propensity scoring • Own the complete ML lifecycle - from exploratory analysis and experimentation all the way through production deployment and operational monitoring • Develop and ship production ML systems spanning self-supervised representation learning, vector similarity search, and supervised classifiers • Leverage distributed computing (Spark/Databricks) and cloud data platforms (AWS, Snowflake) to build and run production ML pipelines at scale • Ensure model quality through rigorous evaluation practices: from embedding validation and retrieval quality to supervised model calibration and production monitoring • Engineer features at scale from demographic, behavioral, and identity data — including handling missing values, encoding strategies, and pipeline-level data quality validation • Contribute ML logic directly into shared production services, working alongside data engineering, software engineering, and product teams