Case Studies | SaaS B2B

Generating risk assessments with AI

LLM Document Creation

About

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

Challenge

The challenge had three focus areas, each essential to a proof of concept the client could build on.

Consistent, structured output

A risk assessment is only useful to a product if it is predictable. The system had to generate valid, structured JSON every time, following a fixed schema with hazards, a five by five likelihood and severity matrix and control measures, with no stray text that would break downstream processing.

Safe, reliable use of the model

Because the service exposes a language model to users, guardrails were needed to prevent misuse. The workflow also had to stay within Amazon Bedrock's invocation limits under real load, handling errors and retries gracefully rather than failing a request.

Compliance-ready deployment and handover

The solution had to be built and deployed with AWS CloudFormation into the client's own AWS account in the London region for data residency, with documentation and training so their team could operate and extend it.

Solution

An user submits a risk assessment request over a WebSocket connection through Amazon API Gateway, passing details such as the assessment title, assessor and activity description. API Gateway forwards the request to an AWS Step Functions state machine that orchestrates the work.

10,000

Requests per month at design scale (projected)

5x5

Risk matrix applied to every generated assessment

Eu-west-2

London region for data residency

By the numbers:

  • 10,000 - Requests per month at design scale (projected)
  • 5x5 - Risk matrix applied to every generated assessment
  • Eu-west-2 - London region for data residency
Changes

The proof of concept met its success criteria: a working system that lets users generate structured risk assessment documents with Amazon Bedrock from a simple set of inputs, deployed into the client's own AWS environment and demonstrated as a suitable base for further development.

  • Assessments from inputs — Users supply an activity description and a few details and receive a full risk assessment with hazards, ratings and control measures.
  • Structured every time — The model returns valid JSON to a fixed schema, so the output can be used directly inside the client's product.
  • Safe by design — Bedrock Guardrails guard against misuse, and the workflow respects invocation limits with waiting and retry logic for reliability.
  • Compliance-aware deployment — The service was deployed with CloudFormation into the client's AWS account in the London region.
  • Handover — Documentation, a demonstration and recorded walkthroughs left the client's team able to operate and extend the service.

With a proven, serverless workflow in place, the client has a foundation it can extend to further document types and richer assessments, moving from a proof of concept towards AI-assisted compliance at scale.

AWS Stack

Amazon API Gateway

(WebSocket) for the live connection that carries requests and responses.

AWS Step Functions

For orchestrating the workflow, including invocation-limit checks and retries.

AWS Lambda

For serverless compute and delivering responses back to the user.

Amazon DynamoDB

For tracking each query, its status and its result.

AWS CloudFormation

For repeatable deployment into the client's own AWS account.

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.