Amazon Augmented AI (A2I)

Human review loops for low-confidence machine learning predictions.

Provider: aws
Kind: Workflow engine
Category: Compute
Official docs: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_HumanLoopConfig.html

Properties · 3 set by this component

property	type	values	required	kind default	this component
compensation	boolean	—	no	false	inherited
customerManagedKey · component property	boolean	—	no	false	inherited
durable	boolean	—	no	true	▸ true
humanReview	boolean	—	no	false	▸ true
integrationSource · component property	enum	rekognition \| textract \| custom	no	—	inherited
managed	boolean	—	no	false	▸ true
pattern	enum	saga \| tcc \| two-phase-commit \| dag	no	dag	inherited
stepsIdempotent	boolean	—	no	false	inherited
taskTimeLimitSeconds · component property	integer	—	no	3600	inherited
tasksPerObject · component property	integer	—	no	—	inherited
workforceType · component property	enum	private \| vendor \| public	no	—	inherited

Rules that apply · 1

Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.

Stateful node without zone redundancyerror

stateful-node-without-zone-redundancy

A stateful compute node pinned to exactly one availability zone disappears with that zone, and its state goes with it. Nodes inside a cell are exempt, because there the cell is deliberately single-zone and redundancy comes from the other cells.

— all of these hold:
— the node's `stateless` is false
— the node sits inside exactly 1 availability-zone group
— this does not hold:
— the node sits inside at least one cell group

Use it in a design

nodes:
  - { id: "augmented-ai", component: "aws-augmented-ai" }

Other Workflow engine products · 44

CloudFlowalibaba-cloudflowalibabacloudworkflow-engine CloudOps Orchestration Servicealibaba-oosalibabacloudworkflow-engine Amazon Managed Workflows for Apache Airflowaws-mwaaawsworkflow-engine AWS Systems Manager Incident Manager and Change Manageraws-ssm-incident-and-change-managerawsworkflow-engine AWS Step Functionsaws-step-functionsawsworkflow-engine Amazon Simple Workflow Serviceaws-swfawsworkflow-engine Azure Durable Functionsazure-durable-functionsazureworkflow-engine Azure Logic Appsazure-logic-appsazureworkflow-engine Cloudflare Workflowscloudflare-workflowscloudflareworkflow-engine Apache Airflowairflowworkflow-engine Argo Workflowsargo-workflowsworkflow-engine Astronomer Astroastronomerworkflow-engine Cadencecadenceworkflow-engine Camunda 8camundaworkflow-engine Conductorconductorworkflow-engine Dagsterdagsterworkflow-engine Digdagdigdagworkflow-engine Flowableflowableworkflow-engine Inngestinngestworkflow-engine jBPMjbpmworkflow-engine Kestrakestraworkflow-engine Luigiluigiworkflow-engine n8nn8nworkflow-engine Node-REDnode-redworkflow-engine Prefectprefectworkflow-engine Prefect Cloudprefect-cloudworkflow-engine Restaterestateworkflow-engine Temporaltemporalworkflow-engine Temporal Cloudtemporal-cloudworkflow-engine Trigger.devtrigger-devworkflow-engine Windmillwindmillworkflow-engine Workatoworkatoworkflow-engine Generic workflow engineworkflow-engineworkflow-engine Zapierzapierworkflow-engine Application Integrationgoogle-application-integrationgcpworkflow-engine Cloud Composer (Managed Service for Apache Airflow)google-cloud-composergcpworkflow-engine Workflowsgoogle-cloud-workflowsgcpworkflow-engine Orchestration Pipelinesgoogle-orchestration-pipelinesgcpworkflow-engine IBM App Connectibm-app-connectibmworkflow-engine Oracle Integration 3oci-integrationociworkflow-engine

and 4 more → Workflow engine page

Source

1# yaml-language-server: $schema=../../schema/component.schema.json
2# Amazon Augmented AI (A2I): human review loops that route low-confidence model predictions to a
3# workteam and write the reviewed results to S3. In maintenance since 30 June 2026 (SageMaker AI - A2I
4# availability change). Per the HumanLoopConfig API Reference a worker has 3,600 seconds (1 hour) to
5# complete a task by default, and TaskCount (1-3 workers per object) must be stated explicitly.
6schemaVersion: "1"
7type: "component"
8id: "aws-augmented-ai"
9kind: "workflow-engine"
10provider: "aws"
11docs: "https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_HumanLoopConfig.html"
12aliases: ["A2I"]
13defaults:
14  managed: true
15  humanReview: true
16  durable: true
17properties:
18  workforceType:
19    type: "enum"
20    values: ["private", "vendor", "public"]
21  taskTimeLimitSeconds:
22    type: "integer"
23    default: 3600
24  tasksPerObject:
25    type: "integer"
26  integrationSource:
27    type: "enum"
28    values: ["rekognition", "textract", "custom"]
29  customerManagedKey:
30    type: "boolean"
31    default: false

The same page as markdown, for an agent: /catalogue/components/aws-augmented-ai.md