{
  "id": 10480852,
  "title": "Automating Amazon Textract adapter lifecycle management across accounts",
  "url": "https://urgent.news/2026/09/28/automating-amazon-textract-adapter-lifecycle-management-across",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-28T15:49:04.000Z",
  "source": {
    "name": "AWS Machine Learning",
    "slug": "aws-machine-learning",
    "url": "https://aws.amazon.com/blogs/machine-learning/automating-amazon-textract-adapter-lifecycle-management-across-accounts/"
  },
  "original_language": "en",
  "account": "The article discusses automated management of Amazon Textract adapter lifecycle across accounts. Amazon Textract is a managed ML service that extracts text, handwriting, layout and structured data from documents. It's used to automate workflows like invoice processing and identity verification. Custom Queries adapters allow fine-tuning extraction for specific document types.\n\nOnce in production, three main challenges emerge:\n1) Moving trained adapters from training to production across accounts, requiring AWS Support tickets.\n2) Determining the right adapter for each document, as Amazon Textract supports only one adapter per AnalyzeDocument API call per page per feature type.\n3) Meeting production security requirements like encryption, network isolation, IAM least-privilege, audit logging and compliance certifications.\n\nThe solution proposed uses a multi-stage processing pipeline with separate concerns for document classification, adapter selection and extraction. Documents are ingested into an encrypted S3 bucket. Pre-classification uses DetectDocumentText to extract raw text and scan for text markers to identify the document version. Adapter selection retrieves the correct adapter ID from AWS Systems Manager Parameter Store. The AnalyzeDocument API is then called with the selected Custom Queries adapter. Results flow to downstream systems.\n\nFor production, API calls are routed through AWS PrivateLink for network isolation and IAM enforces least-privilege access. AWS CloudTrail provides API auditing and Amazon CloudWatch handles monitoring. The architecture decouples adapter management from application logic, allowing adapter updates without code changes or redeployment. Multi-environment topology is recommended for production, with four environments - training, dev, test, and production - each in their own AWS account. Smaller teams can start with two environments within a single account.",
  "summary": "Learn how to operationalize Amazon Textract Custom Queries adapters for production: infrastructure as code with AWS CloudFormation and Terraform, a cross-account adapter promotion process, a pre-classification routing pattern for multiple form versions, and production security controls such as VPC endpoints, encryption, and least-privilege IAM.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}