Case Studies | SaaS B2B

Smarter email decisions, grounded in policy

GraphRAG Email Filtering

About

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

Challenge

The client and Cloud Combinator concentrated on three areas: contextual accuracy of decisions, fully hands-off automation, and cost-effective scalability.

Contextual accuracy

The existing LLM could classify email but had no structured way to draw on the client's policy corpus at decision time, which limited the consistency and defensibility of its verdicts. The project needed to supply deep, policy-derived context to every evaluation.

Hands-off automation

Incoming email had to be processed end-to-end without manual intervention, and the knowledge base behind those decisions had to stay current as policy documents changed, without an operator having to intervene.

Scalable, cost-effective foundations

The architecture had to lean on serverless services and usage-based pricing so it could absorb rising volume while staying within budget, and be deployable and reproducible as code.

Solution

Cloud Combinator delivered the build in clear stages, provisioning every component with AWS CloudFormation so the whole environment could be recreated consistently in the client's AWS account.

100,000

Emails per month processed automatically (design target)

$2,256

Projected monthly AWS cost (MRR)

2

AWS regions for resilient data backup

By the numbers:

  • 100,000 - Emails per month processed automatically (design target)
  • $2,256 - Projected monthly AWS cost (MRR)
  • 2 - AWS regions for resilient data backup
Changes

Acceptance was structured around sign-off of the project's success criteria, confirmation of the internal the client's tech lead's competence with the environment, and a demonstration that the AWS foundation is suitable for continued development of the service. The figures below are projected from the SoW cost model.

  • Policy-aware decisions — The GraphRAG layer supplies graph-derived policy context to the LLM, improving the consistency of hold and release verdicts.
  • Full automation — Amazon SQS and AWS Lambda process incoming email end-to-end with no manual intervention.
  • Always-current knowledge — Scheduled CloudWatch and Lambda refreshes keep the Neptune vector store aligned with the latest policy documents.
  • Resilient, cost-efficient storage — S3 lifecycle rules tier cold data to Glacier and cross-region replication protects the knowledge base.
  • Handover — The entire environment ships as CloudFormation infrastructure-as-code, accompanied by documentation and Loom walkthroughs.

The solution establishes the start of the client's AI capability on AWS. With the GraphRAG foundation in place, the client can pursue further development towards a production-grade email security service, with Cloud Combinator able to guide the steps required to get there.

AWS Stack

Amazon Bedrock

For hosting the embedding models and knowledge base that turn policy documents into semantic context.

Amazon Neptune

For graph-based vector storage and real-time similarity retrieval.

Amazon S3

For raw policy document storage with lifecycle tiering to Glacier and cross-region backup.

AWS Lambda

For automated embedding refresh and email processing without manual intervention.

Amazon CloudWatch

For scheduling knowledge base syncs so decisions reflect current policy.

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.