Case Studies | SaaS B2B

A secure, multi-tenant way to chat with your documents

Intelligent Document Query Platform

About

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

Challenge

The engagement had three focus areas.

Grounded, tuneable document querying

Users needed to query their documents and get answers grounded in that content, with control over model behaviour such as temperature and tone exposed simply through a front end rather than buried in configuration.

Identity-aware, permissioned access

Each customer's data had to stay isolated and access controlled. The model needed to know who was asking without the user stating it, so identity had to flow from Amazon Cognito into the query itself, and results had to respect each user's permissions.

Repeatable, safe provisioning

The client needed to stand up a new customer account containing all the required services in a repeatable way, and to keep the model safe from misuse through pre-screening and guardrails before any query reached the knowledge base.

Solution

Delivered under Cloud Combinator's AI and ML Project Accelerator, the build split into a querying component and a new customer set up component, tested against example data provided by the client.

Bedrock

Grounded document querying with Claude 3

1,000

Scale the solution is costed to support (projected)

CloudFormation

Repeatable per-customer account provisioning

By the numbers:

  • Bedrock - Grounded document querying with Claude 3
  • 1,000 - Scale the solution is costed to support (projected)
  • CloudFormation - Repeatable per-customer account provisioning
Changes

The proof of concept demonstrated both halves of the client's vision: accurate, safe querying over a Bedrock Knowledge Base, and a repeatable, Cognito-permissioned way to provision a new customer, all deployed as infrastructure as code.

  • Querying provenA Bedrock Knowledge Base answering questions grounded in the client's own documents, with temperature and tone tuneable through the front end.
  • Identity-aware by designAmazon Cognito identity flowed into each prompt so the model knows who is asking, with per-customer permissioned access to the data.
  • Safe by defaultA Claude 3 pre-screening stage and guardrails that check every prompt for misuse before it reaches the knowledge base.
  • Repeatable provisioningCloudFormation plus AWS Control Tower to stand up a new customer account with the required services.
  • HandoverA working front end, demonstration and documentation delivered, with the client's technical lead engaged to continue development in Amazon Bedrock.

With the querying capability and the account model both proven, the client have a clear, secure blueprint for taking their document query product to market, on an AWS foundation ready to support continued development.

AWS Stack

Amazon Bedrock Knowledge Bases

For grounded document querying with Claude 3.

Amazon Cognito

For user authentication and identity-aware, permissioned access.

AWS Lambda

And Amazon SQS for the secure, buffered serverless query pipeline.

Amazon S3

For document storage underpinning the knowledge base.

AWS CloudFormation

And AWS Control Tower for repeatable per-customer account provisioning, with AWS Amplify for the front end.

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.