Case Studies | FinTech

Turning dense documents into precise questions

AI Question Mapping

About

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

Challenge

The engagement focused on three areas: extracting precise questions from unstructured documents, aligning generated questions with existing standards, and doing so within a secure and reliable AWS environment.

Decomposing documents into atomic questions

Due diligence documents are long and dense. Transforming them by hand into a set of standalone, unambiguous questions is slow and inconsistent. The client needed a LLM-driven way to reliably break documents down into atomic questions or statements that each capture a single piece of information.

Accuracy and alignment with existing standards

The generated questions had to be accurate and relevant, aligning with the client's existing bank of standard questions rather than drifting into new or off-target phrasing. Quality here is what makes the output usable for real due diligence rather than a novelty.

Security and reliable processing

Given the sensitive nature of the data, the solution had to run inside a secure AWS environment with proper permissions and comply with relevant data protection requirements, while handling model invocation reliably at the volumes involved.

Solution

Cloud Combinator delivered the work through the five-stage Cloud Accelerator programme, taking the client from a shared understanding of the product through to a working delivery in the client's own AWS account, with security best practices and hands-on enablement built in along the way.

Changes

Acceptance was based on signing off the Cloud Accelerator scope of delivery, confirming the technical competence of the client's internal tech lead, and demonstrating that the wider AWS environment is a suitable foundation for the client's further product development. The engagement delivered a functional, Bedrock-based application that turns unstructured due diligence documents into precise, standards-aligned questions.

  • Automated question generation — Unstructured due diligence documents are decomposed into atomic, standalone questions and statements, removing slow and inconsistent manual effort.
  • Standards-aligned output — Generated questions are designed to be accurate, relevant and aligned with the client's existing bank of standard questions.
  • Secure by design — The capability runs on Amazon Bedrock inside the client's own secure AWS environment, keeping sensitive data under the client's control and supporting data-protection compliance.
  • Reliable processing — Batch-style Bedrock invocation with SQS-based response handling keeps the pipeline dependable as document volumes grow.
  • Handover — A hands-on immersion day and best-practice foundation left the client's internal team able to operate and extend the solution independently after sign-off.

With a working question-mapping capability running natively on Amazon Bedrock, the client has a foundation it controls for streamlining due diligence at scale, refining prompts and question quality over time, and reusing the same secure AWS pattern across further AI-driven services.

AWS Stack

Amazon API Gateway

For accepting documents from the client's application and integrating the solution into the client's existing pipeline.

Amazon Bedrock

For the Large Language Model work that decomposes documents into atomic, standards-aligned questions.

AWS Lambda

For invoking Bedrock models with batch-style processing to keep generation reliable under load.

Amazon SQS

For buffering and delivering sets of model responses to reduce exception-handling performance issues.

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.