AWS wellarchitected documentation change
Summary
Rewrote cloud training requirements to specifically include generative AI cost optimization topics like token management and prompt engineering
Security assessment
Focuses exclusively on cost optimization training for generative AI (model selection, token efficiency). No security controls, vulnerabilities, or security features are mentioned in the changes.
Diff
diff --git a/wellarchitected/latest/financial-services-industry-lens/fsicost01.md b/wellarchitected/latest/financial-services-industry-lens/fsicost01.md index e3fdd3f46..a5db06f8f 100644 --- a//wellarchitected/latest/financial-services-industry-lens/fsicost01.md +++ b//wellarchitected/latest/financial-services-industry-lens/fsicost01.md @@ -15 +15 @@ A company-sponsored cloud training program exists, and is required for all cloud - * Digital training: On-demand courses so your team can learn about the latest services when and where it's convenient. +A company-sponsored cloud training program exists, and is required for all cloud stakeholders regardless of their seniority or affiliated organization. For organizations implementing generative AI workloads, this training should include: @@ -17 +17 @@ A company-sponsored cloud training program exists, and is required for all cloud - * Classroom training: In-person and virtual training from instructors who teach your team in a hands-on learning environment. For new employees, it should be part of their onboarding training, and should be mandatory training on a yearly basis for all existing employees and contractors. + * On-demand courses for teams to learn about cost optimization for both traditional cloud services and generative AI workloads. This includes model selection strategies, token optimization, vector store management, and efficient prompt engineering practices. @@ -19 +19,5 @@ A company-sponsored cloud training program exists, and is required for all cloud - * AWS Certification: Validate technical skills and cloud expertise to grow your career and business. + * In-person and virtual training from instructors who teach your team in a hands-on learning environment about cost-effective implementation of generative AI solutions. For new employees, this should be part of their onboarding training, and should be mandatory training on a yearly basis for all existing employees and contractors. + + * Technical skills and cloud expertise, including generative AI implementation, to grow your career and business. Encourage specialization in AI/ML cost optimization paths where available. + + * Prompt engineering for cost, response size control, retrieval scope limiting, embeddings batch sizing, and model A/B testing for price‑performance.