Model Lifecycle and Drift Control: Build and operate the deployment path, including real-time and batch inference, result storage, performance monitoring, drift detection, retraining triggers, feature change management, and rollback, ensuring training and production feature pipelines are provably identical.
Source Integration and Incremental Processing: Implement reliable ingestion from databases, files, APIs and event streams using full-load, incremental, change-data-capture and replayable processing patterns appropriate to each source.
Data Pipeline Architecture and Delivery: Lead the design and implementation of maintainable batch and streaming pipelines that ingest, validate, transform and publish structured, semi-structured and unstructured data across AWS and Databricks.
Technical Programming Skills: Proficient in Python programming language with experience in data manipulation libraries (pandas, numpy), data pipeline frameworks, and API development. Strong SQL skills for data querying, transformation, and database management across various database platforms.