Safety & Governance

Guardrails, data security, responsible AI — 20%

• Guarding Inputs and Outputs — Safety as an engineered layer: Bedrock Guardrails in both directions, custom moderation workflows, hallucination reduction with grounding and structured outputs, prompt-injection and jailbreak defense, and adversarial testing. (3 lessons, 60 min)
• Data Security and Privacy for GenAI — Keeping the data honest while the model uses it: VPC endpoints and IAM for private model access, Lake Formation permissions, PII detection with Comprehend and Macie, masking and anonymization, and retention design. (3 lessons, 50 min)
• Governance and Responsible AI — Proving the system behaves: model cards and lineage, CloudTrail audit trails, drift and misuse detection, transparency through reasoning displays and agent tracing, fairness evaluation with LLM-as-a-judge, and policy-aligned guardrails. (3 lessons, 55 min)