ML/Feature Store Engineer (Advanced Analytics)

The ML/Feature Store Engineer will design and implement the feature engineering pipelines that feed the L2 and L3 Lighthouse AI Agents. This role focuses on computing derived metrics, risk scores, and predictive features from the Medallion Lakehouse and serving them with low latency via an enterprise Feature Store.

The ML/Feature Store Engineer will develop and maintain the enterprise Feature Store (e.g., Feast, Databricks Feature Store) to serve both batch and real-time features to the AI agents. They will collaborate with data scientists to operationalize machine learning models, ensuring that features like supplier risk scores, EAC variance predictions, and skills gap metrics are computed reliably. Building automated pipelines to extract features from the Gold layer of the Lakehouse and the Enterprise Knowledge Graph is a primary responsibility. They will also implement feature monitoring to detect data drift and ensure the ongoing accuracy of the models. Working closely with the API & Security Engineer is necessary to expose these features securely to the agent tier.

Candidates must have 4+ years of experience in data engineering or machine learning engineering. Hands-on experience with Feature Store technologies (Feast, Databricks Feature Store, or similar) is required. Strong proficiency in Python, SQL, and Apache Spark is essential. Experience operationalizing ML models (MLOps) and building scalable data pipelines is expected. Familiarity with graph databases and extracting features from graph structures is highly desired. A solid understanding of data quality monitoring and drift detection techniques is necessary.