Data & ML Engineer

Use sources, schema, transformations, orchestration, idempotency, quality checks, lineage, and monitoring.

Use problem type, data, features, model selection, metrics, validation, deployment, and monitoring.

Use ingestion, chunking, embeddings, retrieval, reranking, citations, hallucination controls, and evals.

Use model registry, deployment, monitoring, drift, data/version lineage, CI/CD, and rollback.

Use eval dataset, scoring rubrics, qualitative review, regression tests, safety cases, and failure analysis.