Official Technical Resources: AWS
Official Technical Resources · AWS. Bedrock, SageMaker and cloud AI documentation. This source has 10 entries admitted under this category's review policy; each links directly to the original.
Authority tiers in this category: A1–A2.
Amazon Bedrock User Guide
Overview of Bedrock's foundation model access, generative AI capabilities, security and application building.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "Bedrock's foundation model access, generative AI capabilities, security and application building"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25
Automate tasks with Agents for Amazon Bedrock
Explains how agents orchestrate models, action groups, knowledge bases and execution traces.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "how agents orchestrate models, action groups, knowledge bases and execution traces"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25
Knowledge Bases for Amazon Bedrock
Introduces data ingestion, vector storage, retrieval, generation and citation configuration.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "data ingestion, vector storage, retrieval, generation and citation configuration"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25
Guardrails for Amazon Bedrock
Explains content filters, denied topics, sensitive information and word policies.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "content filters, denied topics, sensitive information and word policies"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25
Evaluate model performance in Amazon Bedrock
Introduces automatic and human model evaluation, task types, metrics and evaluation jobs.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "automatic and human model evaluation, task types, metrics and evaluation jobs"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25
Prompt management for Amazon Bedrock
Explains prompt templates, variables, versions, variants and test management.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "prompt templates, variables, versions, variants and test management"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25
Amazon SageMaker AI Developer Guide
Overview of data preparation, training, tuning, deployment, MLOps and governance.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "data preparation, training, tuning, deployment, MLOps and governance"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25
SageMaker JumpStart
Introduces pretrained models, solution templates, fine-tuning and rapid deployment.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "pretrained models, solution templates, fine-tuning and rapid deployment"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25
AWS Neuron documentation
Explains compilation, training, inference and performance tuning on Trainium and Inferentia.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "compilation, training, inference and performance tuning on Trainium and Inferentia"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25
Machine Learning Lens
Reviews ML workloads for operational excellence, security, reliability, performance, cost and sustainability.
Evidence use: Directly supports the vendor's product mechanisms, interfaces and official recommendations covered by "reviewing ML workloads for operational excellence, security, reliability, performance, cost and sustainability"; conclusions beyond that product scope need cross-checking with independent sources.
Limitations: Directly supports facts only about the vendor's current public products, interfaces or methods; it cannot replace independent cross-platform comparison. Online documentation changes over time; recheck the original page for versions, pricing, limits and regional availability.
Link last checked: 2026-07-25