Machine Learning Engineer II

ML Development & Modeling

Build and maintain end-to-end ML pipelines for training, evaluation, and batch inference across use cases such as identity resolution, audience segmentation, and content affinity modeling.

Implement and experiment with supervised, unsupervised, and ranking models in Python (scikit-learn, XGBoost/LightGBM, PyTorch).

Engineer features from first-party viewership, engagement, subscription, and behavioral signals, guarding against data leakage, collinearity, and training/serving skew.

Run structured offline experiments; evaluate with the right metrics (precision/recall, F1, AUC-ROC, calibration, lift) and document findings in MLflow.

ML Infrastructure & Engineering

Develop and maintain data and feature pipelines on Databricks (PySpark, Delta, Workflows) that feed the feature store and model-training workflows, with attention to idempotency and reproducibility.

Write clean, tested, production-quality Python following engineering best practices (unit tests, code reviews, CI/CD).

Use MLflow for experiment tracking, model registration, and versioning under the guidance of senior engineers.

Support deployment and monitoring of batch inference jobs integrated with downstream activation platforms (e.g., Mosaic, FreeWheel, GAM) and data in Snowflake.

Agentic AI & Modern Tooling

Use AI-assisted development tools (Cursor, GitHub Copilot, Amazon Q) to accelerate coding, debugging, and documentation under guidance.

Leverage Databricks Genie for natural-language exploration of governed Unity Catalog datasets — querying ML feature tables, model outputs, and audience segments.

Use Snowflake Cortex (Copilot / Cortex Analyst) to accelerate data analysis and SQL authoring against audience and identity schemas. Learn and apply prompt-engineering patterns for LLM-assisted data exploration and feature generation, and participate in evaluating MCP (Model Context Protocol) tooling as the team expands agentic workflows.

Collaboration & Growth

Partner with Senior and Staff MLEs to understand system-design decisions and contribute meaningfully to technical discussions.

Work cross-functionally with Data Engineering, Feature Engineering, and Analytics to ensure data quality and pipeline reliability.

Document models, pipelines, and experiments clearly for team knowledge sharing.