Case Studies | FinTech

One assistant for cost of living, health and careers

IW Build - Staff Support Knowledge Base

About

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

Challenge

The challenge had three focus areas that the build needed to address.

Trusted answers across three topics

Users needed accurate support on cost of living, health and wellbeing, and careers, grounded in the client's own documentation rather than open-ended model output that might stray off-topic.

Fair, tiered access with cost control

Giving every user access to a large language model raises cost and fairness questions. The service had to track usage per user and enforce limits according to their subscription level.

Understanding what users need

Beyond answering questions, the client wanted to identify the topics being discussed and interrogate that data, so they needed a way to categorise conversations and query them in natural language.

Solution

The client's guidance documents are held in an Amazon S3 bucket that feeds an Amazon Bedrock Knowledge Base backed by an Amazon OpenSearch Serverless vector store, with uploads and deletions automatically re-syncing the knowledge base. Users chat through a WebSocket API on Amazon API Gateway; an AWS Lambda logs the conversation in Amazon DynamoDB and checks the user's token usage against their subscription tier before an AWS Step Functions workflow passes the request to Amazon Lex and an Amazon Bedrock Agent. Amazon Bedrock Guardrails keep the assistant on-topic and protected from misuse.

3

Support topics: cost of living, health and wellbeing, and careers

24/7

Always-on chat support grounded in the client's own guidance

100%

Deployed as code with AWS CloudFormation

By the numbers:

  • 3 - Support topics: cost of living, health and wellbeing, and careers
  • 24/7 - Always-on chat support grounded in the client's own guidance
  • 100% - Deployed as code with AWS CloudFormation
Changes

Cloud Combinator delivered a working AI support assistant that answers the client's users' questions across three topic areas, governs usage by subscription tier, and turns conversations into queryable insight, all deployed into the client's own AWS account as infrastructure as code.

  • Grounded support — Users get answers drawn from the client's own documentation across cost of living, health and wellbeing, and careers.
  • Tiered and governed — Per-user token tracking in Amazon DynamoDB enforces usage limits by subscription level, keeping cost and access fair as adoption grows.
  • Insight from conversations — A nightly categorisation job tags conversations by topic, and a natural-language query pipeline lets the client interrogate what users are asking about.
  • Safe and on-topic — Amazon Bedrock Guardrails and prompt engineering keep the assistant scoped to its intended purpose and protected from misuse.
  • Handover — Cloud Combinator delivered the solution into the client's AWS account with a demonstration, materials and documentation.

With the assistant live, the natural next step (projected) is to connect it to the client's production front end, add a hallucination check for extra assurance, and extend the service to new markets and regions as the user base grows.

AWS Stack

Amazon Bedrock Knowledge Bases

And Agents for grounded, conversational support.

Amazon OpenSearch Serverless

For the vector store powering retrieval.

Amazon Lex

For natural-language conversation flow.

Amazon Bedrock Guardrails

For on-topic, safe responses.

AWS Lambda

And AWS Step Functions for orchestration and usage-quota management.

Amazon DynamoDB

For conversation logs, token tracking and tier-based limits.

Amazon API

Gateway, including a WebSocket API, for real-time chat.

Amazon EventBridge

For the scheduled nightly conversation categorisation.

Amazon S3

With cross-region replication for resilient document storage.

Amazon VPC

And AWS CloudFormation for a secure network and repeatable infrastructure as code.

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.