--- title: Build and Run an Intelligent Document Processing (IDP)... description: A walkthrough for building an Intelligent Document Processing (IDP) pipeline on AWS that automates email ingestion, document classification, and PII... canonical: https://daily.dev/posts/build-and-run-an-intelligent-document-processing-idp-system-in-the-cloud-zhylmyuz6 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Build and Run an Intelligent Document Processing (IDP) System in the Cloud | daily.dev og:description: A walkthrough for building an Intelligent Document Processing (IDP) pipeline on AWS that automates email ingestion, document classification, and PII... og:url: https://daily.dev/posts/build-and-run-an-intelligent-document-processing-idp-system-in-the-cloud-zhylmyuz6 og:image: https://api.daily.dev/og/posts/ZhYlmyuz6.png og:image:alt: Build and Run an Intelligent Document Processing (IDP) System in the Cloud og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Build and Run an Intelligent Document Processing (IDP) System in the Cloud **[Towards Data Science](https://daily.dev/sources/tds)** · 14 min read · 0 upvotes · 0 comments ## Summary A walkthrough for building an Intelligent Document Processing (IDP) pipeline on AWS that automates email ingestion, document classification, and PII extraction. The system uses Gmail OAuth credentials stored in AWS Secrets Manager, downloads image attachments to S3, runs Amazon Textract for OCR, and passes the extracted text to Claude Sonnet 4.6 via AWS Bedrock to classify documents (passport, driving licence, bank statement) and extract PII fields. Three Lambda functions handle each stage, orchestrated by a Step Function triggered on a cron schedule via EventBridge. The article includes sample outputs, environment variable configs, and notes on what would be needed to productionize the system. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can I use AWS Textract and Bedrock together to extract PII from scanned documents like passports and bank statements? Amazon Textract performs OCR on document images to extract raw text and confidence scores, then that text is passed to Amazon Bedrock running Claude Sonnet 4.6 to classify the document type and extract structured PII fields such as first name, last name, date of birth, and address. The pipeline stores intermediate JSON outputs in S3 folders named raw, textract, and results, orchestrated by a Step Functions workflow triggered daily via EventBridge. _daily.dev surfaces practical AWS pipeline patterns for teams automating document intake and PII extraction._ ### Do I need Textract if I'm using Bedrock Claude Sonnet to read documents? Not necessarily for simple, high-quality images: Bedrock's Claude Sonnet model can classify and extract information directly from clean images without an OCR step. Textract becomes valuable when inputs are more complex, such as PDFs, handwritten text, or lower-quality scans, where a dedicated OCR pass improves reliability before handing text to the language model for classification and extraction. _Comparing OCR-plus-LLM versus LLM-only pipelines is easier with real-world tradeoffs surfaced on daily.dev._ ## Similar posts on daily.dev - [Automate Your Work with Intelligent Document Processing](https://daily.dev/posts/automate-your-work-with-intelligent-document-processing-0elxpdfc7) · Netguru · 0 upvotes · 0 comments - [How Ricoh built a scalable intelligent document processing solution on AWS](https://daily.dev/posts/how-ricoh-built-a-scalable-intelligent-document-processing-solution-on-aws-48ujfcjyy) · AWS · 0 upvotes · 0 comments - [Programmatically creating an IDP solution with Amazon Bedrock Data Automation](https://daily.dev/posts/programmatically-creating-an-idp-solution-with-amazon-bedrock-data-automation-ii93woz6i) · AWS · 2 upvotes · 0 comments - [Building with Databricks Document Intelligence and Lakeflow](https://daily.dev/posts/building-with-databricks-document-intelligence-and-lakeflow-apnw3dkdv) · databricks · 0 upvotes · 0 comments - [Building a Content Generation Pipeline with DigitalOcean Serverless Inference](https://daily.dev/posts/building-a-content-generation-pipeline-with-digitalocean-serverless-inference-xui9o3ycj) · DigitalOcean Community · 1 upvotes · 0 comments --- Tags: [#aws](https://daily.dev/tags/aws), [#aws-lambda](https://daily.dev/tags/aws-lambda) [View this post on daily.dev](https://daily.dev/posts/build-and-run-an-intelligent-document-processing-idp-system-in-the-cloud-zhylmyuz6) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Build and Run an Intelligent Document Processing (IDP) System in the Cloud","url":"https://daily.dev/posts/build-and-run-an-intelligent-document-processing-idp-system-in-the-cloud-zhylmyuz6","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/build-and-run-an-intelligent-document-processing-idp-system-in-the-cloud-zhylmyuz6"},"datePublished":"2026-07-24T15:41:26.064Z","dateModified":"2026-09-13T20:55:03.330Z","description":"A walkthrough for building an Intelligent Document Processing (IDP) pipeline on AWS that automates email ingestion, document classification, and PII...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/2bd2e2c66158eac61a780a85752f7e10?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/2bd2e2c66158eac61a780a85752f7e10?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Towards Data Science","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Towards Data Science","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/tds","url":"https://daily.dev/sources/tds"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/build-and-run-an-intelligent-document-processing-idp-system-in-the-cloud-zhylmyuz6","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"aws,aws-lambda","timeRequired":"PT14M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Towards Data Science","item":"https://daily.dev/sources/tds"},{"@type":"ListItem","position":3,"name":"Build and Run an Intelligent Document Processing (IDP) System in the Cloud"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/build-and-run-an-intelligent-document-processing-idp-system-in-the-cloud-zhylmyuz6#faq","mainEntity":[{"@type":"Question","name":"How can I use AWS Textract and Bedrock together to extract PII from scanned documents like passports and bank statements?","acceptedAnswer":{"@type":"Answer","text":"Amazon Textract performs OCR on document images to extract raw text and confidence scores, then that text is passed to Amazon Bedrock running Claude Sonnet 4.6 to classify the document type and extract structured PII fields such as first name, last name, date of birth, and address. The pipeline stores intermediate JSON outputs in S3 folders named raw, textract, and results, orchestrated by a Step Functions workflow triggered daily via EventBridge. daily.dev surfaces practical AWS pipeline patterns for teams automating document intake and PII extraction."}},{"@type":"Question","name":"Do I need Textract if I'm using Bedrock Claude Sonnet to read documents?","acceptedAnswer":{"@type":"Answer","text":"Not necessarily for simple, high-quality images: Bedrock's Claude Sonnet model can classify and extract information directly from clean images without an OCR step. Textract becomes valuable when inputs are more complex, such as PDFs, handwritten text, or lower-quality scans, where a dedicated OCR pass improves reliability before handing text to the language model for classification and extraction. Comparing OCR-plus-LLM versus LLM-only pipelines is easier with real-world tradeoffs surfaced on daily.dev."}}]} ```