Case Studies | Healthcare & Life Sciences

A secure natural-language query service for care records, built on AWS

LLM Report Query Chatbot

About

A Healthcare & Life Sciences business working with Cloud Combinator on AWS. The client is anonymised at their request.

Challenge

The build had three focus areas.

Enforcing access at the data boundary

Different care homes hold different records, and different professionals are entitled to different subsets. The system had to guarantee a caretaker could query only their authorised homes, and never retrieve a blocked member, with that rule enforced by the platform rather than the application layer alone.

Accurate retrieval from unstructured records

The service needed to extract genuinely relevant answers from the documents the client provided, converting them into a

Safety against misuse

Because the interface is conversational and points at sensitive data, prompts aimed at misusing the AI had to be stopped before they ever reached the knowledge base, and the entire query pipeline had to run over private connections inside a VPC.

Solution

Cloud Combinator stood the service up in the eu-west-1 region using AWS CloudFormation, structured around a knowledge base per care home and an access model enforced by a Lambda Authorizer and short-lived, attribute-scoped credentials.

Per home

An isolated knowledge base and index for each care home

Role-based

Access enforced by a Lambda Authorizer and scoped STS credentials

Eu-west-1

Pipeline runs privately in the UK region

By the numbers:

  • Per home - An isolated knowledge base and index for each care home
  • Role-based - Access enforced by a Lambda Authorizer and scoped STS credentials
  • Eu-west-1 - Pipeline runs privately in the UK region
Changes

The service is scoped to meet a clear set of acceptance criteria: OpenSearch Serverless indices with per-home access controls, a VPC-private query pipeline, guardrails against misuse,

  • Access enforced at the boundary — A caretaker reaches only their authorised homes, and blocked members are removed from results before any query runs.
  • Conversational retrieval — Natural-language questions return contextually accurate answers drawn from the client's own documents, streamed back in real time.
  • Guarded against misuse — Bedrock Guardrails stop misuse-oriented prompts reaching the knowledge base, and the whole pipeline runs on private VPC connections.
  • Auditable — Conversation history is logged to DynamoDB, giving a record of what was asked and answered.
  • Handover — The service is deployed via AWS CloudFormation with testing, a demonstration and documentation so the client can operate it independently.

Access enforced through a Lambda Authorizer, and accurate natural-language retrieval from the client's documents. Together they define a search experience that is conversational for the user and rigorously controlled underneath.

With a secure, natural-language route into their records, the client's care professionals can find what they need in plain English, while the platform holds the line on exactly who is allowed to see what.

AWS Stack

Amazon Bedrock Knowledge Base

For embedding documents and answering natural-language queries with Claude, protected by Bedrock Guardrails.

Amazon OpenSearch Serverless

For the per-home vector indices that back retrieval.

Amazon API Gateway

For the real-time WebSocket connection to care professionals.

AWS Lambda

For the Lambda Authorizer and the query execution logic.

Amazon DynamoDB

For user access mappings, blocked-member records and chat history.

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.