Skip to Content
Cloud Engineering Academy
16. Generative AI on AWS for Banking

Module 16 cheat sheet

One-page revision summary for Generative AI on AWS for Banking: services, Azure mapping, CLI commands, limits, decision rules, exam traps and cost traps.

AIF-C01AIP-C01

Assumes: AWS CLI v2 · AWS SDK for .NET v4 (AWSSDK.BedrockRuntime) · Region us-east-1 · facts as stated in the Module 16 lessons (checked 2026-11-18). Verified against official sources on 2026-11-18.

A printable summary of Module 16. Each row links back to the lesson with the detail. Model IDs, prices and availability change often: confirm them in the Bedrock console before you rely on them.

Key services

ServiceWhat it is for
Amazon Bedrock (Converse API)Call models from many providers through one request shape; stateless, so history costs tokens each turn
Inference profiles (us., eu., global.)Route a request across Regions; Claude 5.5 models have no in-Region support in us-east-1
Bedrock Knowledge BasesManaged RAG: ingest approved documents, retrieve with citations; managed or customer-managed
Bedrock GuardrailsContent, prompt-attack, denied-topic, PII and grounding checks; ApplyGuardrail works with no model
Amazon Bedrock AgentCoreHosting and governance for agents (Runtime, Gateway, identity, memory, policy, observability)
KiroSuccessor to Amazon Q Developer IDE plugins; Q in the console continues
Bedrock model invocation loggingA second copy of your prompts: classify, encrypt, set retention
Bedrock Evaluations / fmevalEvaluate foundation models, a model-risk validation input

.NET / Azure → AWS

Microsoft / AzureAWS
Azure OpenAI / AI FoundryAmazon Bedrock
Azure.AI.OpenAIAWSSDK.BedrockRuntime (Converse)
Microsoft.Extensions.AI IChatClientSame interface over Bedrock or a local model
Azure AI Search (RAG)Bedrock Knowledge Bases
Content Safety, Prompt ShieldsBedrock Guardrails (prompt attack filter)
Azure AI Agent ServiceBedrock AgentCore
GitHub CopilotKiro

Most useful CLI commands

TaskCommand
List modelsaws bedrock list-foundation-models --profile learn-sandbox
Read a profile's destination Regionsaws bedrock get-inference-profile --inference-profile-identifier <id>
One-off model callaws bedrock-runtime converse --model-id <profile-id> --messages file://msg.json
Create / version a guardrailaws bedrock create-guardrail … · aws bedrock create-guardrail-version --guardrail-identifier <id>
Test text against a guardrailaws bedrock-runtime apply-guardrail …
Delete a guardrail (teardown)aws bedrock delete-guardrail --guardrail-identifier <id>
Sync a knowledge baseaws bedrock-agent start-ingestion-job …
Query a knowledge baseaws bedrock-agent-runtime retrieve …
Check logging configaws bedrock get-model-invocation-logging-configuration
Anything left tagged?aws resourcegroupstaggingapi get-resources --tag-filters Key=Project,Values=cea-lab

Key limits and defaults

ItemValueLesson
Claude 5.5 context / max output1M tokens / 128KBedrock basics
Titan Text Embeddings V2InvokeModel only; 8K-token input; 1,024, 512 or 256 dimensionsBedrock basics
Prompt caching (5.5 cards)512-token minimum, up to 4 checkpoints, 5-minute or 1-hour lifeBedrock basics
Guardrail text unitUp to 1,000 characters; 5,600 characters bills as 6 unitsGuardrails
Guardrail prices per 1,000 unitsContent filters and denied topics 0.15 USD; PII and grounding 0.10 USD; word and regex filters freeGuardrails
Denied-topic definition200 characters (Classic), 1,000 (Standard)Guardrails
Managed KB embedding1,024 dimensions, float32; type cannot change after creationKnowledge Bases
Retention mode aws_reviewUp to 30 days, may be reviewed by AWS; some models require itData privacy

Decision rules

If…Then…
Calling a Claude 5.5 model from us-east-1Use a us. or global. profile ID, not the bare model ID
Personal data and residency matterPrefer a geographic profile or single Region; read destinations with GetInferenceProfile
Choosing a RAG storeManaged Knowledge Base first; customer-managed only for a specific store need
Users see different documentsMetadata filters built server-side from verified identity
Guardrail on a RAG callMark only the question for evaluation; scanning all passages can cost more than the model
Agent can change stateAllow-list tools, step and token limits, human approval for writes
Output affects a customer or moneyHuman review, maker-checker, default is not released
Need to block a modelIAM or SCP deny on bedrock:InvokeModel; access is automatic on first use

Common exam traps

  • Retired Model access page: commercial Regions enable models on first use; the control is IAM and SCPs.
  • Closed services still appear in exam guides: Kendra, Q Business, Clarify, Model Monitor, Ground Truth, A2I (closed 2026-07-30). Replacements: Bedrock Knowledge Bases, Amazon Quick, SHAP and MLflow, Evidently, a self-built review queue.
  • Bedrock Agents Classic is closed to new customers; use AgentCore. Models, Knowledge Bases and Guardrails are unaffected.
  • Guardrails are probabilistic and text-only: tool results, tool definitions and tool-call arguments are not evaluated; the trace holds the unmasked match.
  • Relevance is not permission: similarity search does not enforce access control.
  • RAG is not fine-tuning: RAG adds documents at question time and needs no training.
  • Prompt injection (OWASP LLM01): indirect injection through retrieved content is the main RAG risk.

Cost traps

TrapAvoid it by
Knowledge base left runningPaid while it exists: delete it the same day, or use local pgvector
Guardrail evaluating the whole promptEvaluate the question only; billing follows text units
Long prompts and chat historyHistory is resent each turn; long-context rates apply above a threshold
Provisioned ThroughputBilled hourly even when idle
Unbounded agent loopsCap steps, tokens and time; cap MaxTokens on every call
Invocation logsRetention and storage grow; set lifecycle rules and encrypt

Independent learning resource — not affiliated with, endorsed or sponsored by Amazon Web Services. AWS and related marks are trademarks of Amazon.com, Inc. or its affiliates. All banking data in examples is fictional. Sample architectures are for learning and are not automatically compliant with GDPR, PCI DSS or any banking regulation.

On this page