References: Batch Job Operations, Privacy, and Compliance

1. Batch processing - Wikipedia - Foundational coverage of the asynchronous-job-submission pattern that LLM batch APIs implement at half the synchronous cost.

2. General Data Protection Regulation - Wikipedia - Comprehensive coverage of the EU privacy regime including data subject rights and lawful bases for processing that constrain LLM logging schemas.

3. Health Insurance Portability and Accountability Act - Wikipedia - Coverage of the US healthcare privacy law that drives PHI redaction requirements in healthcare-adjacent LLM applications.

4. Designing Data-Intensive Applications - Martin Kleppmann - O'Reilly - The chapters on batch and stream processing provide the broader systems context for the LLM batch-API material in this chapter.

5. The Privacy Engineer's Manifesto - Michelle Dennedy, Jonathan Fox, and Tom Finneran - Apress - Practical reference for privacy-by-design including the redaction, hashing, and retention strategies this chapter applies to LLM logs.

6. Anthropic Batch API Documentation - Anthropic - Authoritative reference for the Message Batches API including submission format, polling, idempotency, and the 50% discount used in this chapter's worked examples.

7. OpenAI Batch API Documentation - OpenAI - Reference for OpenAI's batch endpoint including JSONL input format, status polling, and discount structure that mirrors Anthropic's design.

8. Google Gemini Batch Mode - Google - Reference for Gemini's batch processing including the discount structure and submission format used in this chapter's cross-vendor comparisons.

9. NIST Privacy Framework - NIST - The US standards body's privacy risk-management framework; useful for engineers building enterprise compliance programs around LLM logging.

10. AICPA SOC 2 Trust Services Criteria - AICPA - The reporting standard most enterprises require for vendor due diligence; understanding it helps engineers design LLM logging that survives a SOC 2 audit.