Case Studies | SaaS B2B

Answering the holiday question

Bedrock Knowledge Base Query

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 an assistant users and the business could trust.

Grounding answers in Sawday's content

Recommendations had to come from Sawday's own information, not a model's guesswork. The first focus was ingesting website content into a vector database within an Amazon Bedrock Knowledge Base so answers were grounded and relevant.

Keeping the knowledge current

Destinations, reviews and availability change constantly. The second focus was atomic updates, so that when a new review is left or a booking changes availability, the knowledge base is updated without disrupting existing data.

Safe, real-time conversation

Users expect an immediate reply, and the business needs the assistant to stay on topic. The final focus was a secure, real-time chat pipeline with Amazon Bedrock Guardrails to prevent misuse and keep responses appropriate.

Solution

We delivered the solution in four phases, deploying everything as infrastructure-as-code with AWS CloudFormation.

Real-time

WebSocket chat with responses grounded in Sawday's content

Guardrails

Bedrock Guardrails keep answers safe and on topic

1,000+

Users sized for in the deployed cost model

By the numbers:

  • Real-time - WebSocket chat with responses grounded in Sawday's content
  • Guardrails - Bedrock Guardrails keep answers safe and on topic
  • 1,000+ - Users sized for in the deployed cost model
Changes

The engagement delivered a grounded, real-time recommendation assistant deployed as infrastructure-as-code in Sawday's AWS environment. It answers natural-language questions about holiday destinations from Sawday's own content, stays current as reviews and bookings

  • Grounded recommendations delivered — A Bedrock Knowledge Base over OpenSearch Serverless answers destination questions from Sawday's own website content.
  • Always current — Atomic updates keep the knowledge base fresh as new reviews and bookings arrive, without disrupting existing data.
  • Safe and real-time — A WebSocket pipeline with SQS and a rate-limiting Step Function returns guardrailed answers to users live.
  • Handover — The solution was deployed via CloudFormation into Sawday's own AWS account, with a demonstration, documentation and a handover session.

Change, and keeps responses safe through Bedrock Guardrails, meeting the project's success criteria.

With the assistant live, the natural next steps (projected) would be to enrich the knowledge base with more of Sawday's content, refine prompts and retrieval for sharper recommendations, and explore multi-agent and analytical capabilities as Sawday's grows the service.

AWS Stack

Amazon OpenSearch Serverless

For the vector database behind the knowledge base.

Amazon API Gateway

For both content ingestion and the real-time WebSocket chat interface.

AWS Step Functions

And Amazon SQS for orchestrating and rate-limiting query processing.

AWS Lambda

For ingestion, connection management and response delivery.

Amazon DynamoDB

For tracking WebSocket connections, and Amazon S3 for content storage.

AWS CloudFormation

For deploying the whole solution as 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.