Case Studies | FinTech

Putting generative AI to work in insurance operations

Generative AI Proof of Concept Suite

About

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

Challenge

The engagement had three focus areas, each a separate proof of concept with its own success measure.

Reviewing contracts faster without losing rigour

The client review large numbers of contracts and wanted to accelerate that work. The need was to read a contract clause by clause, judge each clause against their own compliance criteria, and flag risk consistently, all while explaining the reasoning so a human reviewer could trust and audit the result.

Resolving complex, multi party claims

Some claims have no single source of blame or are highly complex, and resolving them is slow because it takes time to find and review comparable past cases. The challenge was to let an user upload a new claim and a policy, then receive a recommended resolution grounded in similar historical cases rather than in the model's general knowledge alone.

Turning unstructured documents into structured data

Policy documents and claimant email chains hold valuable information locked in free text. The client needed a reliable way to extract key policy and claim fields and, in the case of correspondence, to read the sentiment and turning points of an interaction, all returned as clean structured data ready for downstream use.

Solution

We delivered the three use cases as parallel proofs of concept on a shared Amazon Bedrock foundation, prompt engineering and testing each against the client's own example documents so the outputs reflected their real criteria and language.

3

Generative AI proofs of concept delivered on Amazon Bedrock

3-tier

Clause risk classification: Acceptable, Minor, Major, with rationale

JSON

Structured, auditable output from every build

By the numbers:

  • 3 - Generative AI proofs of concept delivered on Amazon Bedrock
  • 3-tier - Clause risk classification: Acceptable, Minor, Major, with rationale
  • JSON - Structured, auditable output from every build
Changes

All three proofs of concept were delivered and demonstrated against the client's own example documents, giving them concrete, working evidence of where generative AI on AWS can add value across contract review, claims and document processing.

  • Contract review provenClause by clause classification with a reason and compliance reference for each clause, tested across multiple prompt strategies to tune consistency.
  • Claim resolution grounded in precedentRecommendations that determine fault, coverage and deductibles by referencing similar past cases, not just model generalities.
  • Documents made analysableReliable extraction of policy and claim fields plus sentiment analysis of correspondence, delivered as clean structured data.
  • Explainability by designEvery output pairs a decision with its justification, keeping a human reviewer in control and the result auditable.
  • HandoverThe client left the engagement with three working reference implementations and the prompt and architecture patterns behind them, ready to inform a decision on production investment.

With the value of each use case demonstrated on their own material, the client are well placed to prioritise which workflow to take from proof of concept to production, building on a governed Amazon Bedrock foundation that can support whatever they take on next.

AWS Stack

Amazon Bedrock

For governed access to foundation models powering all three use cases.

Amazon Bedrock Knowledge Bases

For retrieval augmented claim resolution grounded in past cases.

Amazon S3

For secure storage of the source contracts, policies, claims and correspondence.

AWS Lambda

For serverless orchestration of the document processing and model calls.

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.

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

See our Privacy Policy for details.