Case Studies | SaaS B2B

Automating document checks with AI

Document Extraction

About

A SaaS B2B business working with Cloud Combinator on AWS. The client is anonymised at their request.

Challenge

The challenge had three focus areas, each essential to a proof of concept the client could trust.

Extracting reliable data from mixed documents

Uploaded files could be scanned images or text-based PDFs, so the pipeline first had to detect the document type and then extract text accurately from either. Without dependable extraction, every downstream decision would be built on shaky ground.

Making consistent, objective decisions

The core requirement was automated approve or reject decisions driven by a set of validation rules applied to the extracted content. The decisions had to be consistent and objective, reducing the variability of manual checking, and they had to be resistant to the model inventing information that was not in the document.

Privacy and full traceability

Because the documents contain personal information, the solution needed controls to prevent sensitive data being exposed, and comprehensive logging so that every upload, extraction and decision could be reviewed for auditing and compliance.

Solution

An user uploads a PDF to Amazon S3 using a secure pre-signed URL generated through Amazon API Gateway. The upload triggers a Lambda function that determines whether the file is a scanned image or a text-based PDF, then starts an asynchronous Amazon Textract job to extract the text. The upload and the Textract job are both logged to Amazon DynamoDB.

22,000

Documents per month at design scale (projected)

2

Bedrock Guardrails: PII filter and contextual grounding

100%

Serverless, event-driven and logged for audit

By the numbers:

  • 22,000 - Documents per month at design scale (projected)
  • 2 - Bedrock Guardrails: PII filter and contextual grounding
  • 100% - Serverless, event-driven and logged for audit
Changes

The proof of concept met its objective: an end-to-end pipeline that automates document processing, produces consistent approve or reject decisions and records every stage for full auditability, all on managed AWS services.

  • Automated extraction — The pipeline detects the document type and uses Amazon Textract to pull text from both scanned images and digital PDFs.
  • Objective decisions — A Bedrock model applies the validation rules to the extracted content and returns a consistent Approved or Rejected outcome.
  • Privacy by design — Guardrails filter sensitive information and ground the decision in the supplied data and rules, protecting personal data and improving reliability.
  • Full traceability — Every upload, extraction and decision is logged to DynamoDB with relevant metadata for auditing, monitoring and debugging.
  • Handover — The pipeline was delivered as a working proof of concept on managed AWS services that the client can evaluate and build on.

With a proven, serverless pipeline in place, the client has a foundation it can extend to further document types and validation rules, moving from a proof of concept towards automated document processing at production scale.

AWS Stack

Amazon S3

For secure document storage via pre-signed upload URLs.

Amazon API Gateway

For the endpoints that generate upload URLs and handle client interaction.

AWS Lambda

For the event-driven compute that orchestrates the workflow.

Amazon Textract

For optical character recognition across scanned and digital documents.

Amazon DynamoDB

For event logging, metadata and end-to-end traceability.

YOU MIGHT LIKE

Related success stories

View all case studies

Case Studies | Insights

Utilising Language Recognition, Speed, and Enhanced Security to Make Social Media a Force for Good

  • Here, we take a detailed look at how the Cloud Combinator team collaborated with another cutting-edge AI service provider that provides intelligent systems to “make social media more social” for brands and users alike.
  • Arwen AI is a UK-based startup specialising in AI solutions to manage and enhance brands’ social media interactions. Founded in 2020 by Matt McGrory, Dr. David Cole, and Joel Bailey, Arwen. AI focuses on using AI to automatically detect and remove spam, toxic comments, and other unwanted content from social media platforms.
  • The team at Arwen have three core products. ‘Moderate’ is focused on identifying and removing toxic content from social media channels. ‘Engage’ helps brands identify and engage with meaningful conversations on social media, and ‘Customize’ allows brands to apply bespoke algorithms to their channels - creating an even more effective moderation and engagement.
Read more
CONTACT US

Ready to turn AI into impact?

We'll help you spot the highest-value opportunities, reduce risk around your first AI initiative, and define a clear path to results from day one.

Why talk to us:

Outcome-driven recommendations

AWS-recognised delivery expertise

Risk-aware AI adoption

Clear next step, not a sales pitch

Start with a focused 20-minute conversation about your goals — no pressure, no commitment.

We use cookies

This website uses cookies to enhance user experience and to analyze performance and traffic on our website.

See our Privacy Policy for details.