Machine Learning
CoreAmazon Bedrock
Managed foundation-model and knowledge-base integrations used when a data pipeline must prepare, store, or retrieve model-ready data.
Key points
- Bedrock integrations can enrich unstructured records, generate embeddings, and support knowledge-base ingestion or retrieval.
- Validate generated structured output and preserve source provenance; keep deterministic pipeline checks outside the model.
Best-known use cases
- Generate embeddings or enrich unstructured records with foundation models.
- Prepare enterprise content for retrieval-augmented generation knowledge bases.
What candidates often confuse it with
- Bedrock provides foundation-model and knowledge-base integration; the pipeline still owns data preparation, storage, schema, and governance.
Key takeaway
Use Bedrock only for a bounded model-assisted step within an otherwise validated data-engineering workflow.
Relevant exam tasks
- D1.2 — Task 1.2: Transform and process data
- 1.2.10 — Integrate large language models (LLMs) for data processing.
- D2.4 — Task 2.4: Design data models and schema evolution
- 2.4.6 — Describe vectorization concepts (for example, Amazon Bedrock knowledge base).