Launching an AWS skeleton to modernise fraud detection services for Comsec

  • Customer
  • Industry
  • Service
  • Segment
  • Author
  • Comsec
  • Professional Services
  • AWS Adoption & ML
  • SMB
  • Jodie Rhodes
Explore more
At a glance

Comsec works with brands across industries – fashion, retail, luxury, media, sports, finance, and legal – to detect theft and fraud in physical and digital sales.

Challenge

Comsec needed to modernise their fraud detection but lacked in-house AWS expertise.

Solution

Firemind created an AWS infrastructure skeleton for Comsec, which could be further developed with machine learning capabilities.

Services used
  • AWS Step Functions
  • AWS Key Management Service (KMS)
  • Amazon S3
  • AWS Lambda
Outcomes
  • AWS infrastructure built, tested, and launched in 25 days.
  • Extensive training and documentation provided by Firemind.
Business challenges

Tackling Fraud prevention

Whilst having a very specialised team, Comsec did not have the inhouse capabilities to run with turnkey solutions specifically within AWS. They were in need of a specialist team with an AWS centric skillset.

Some of their client base consist of worldwide retail brands, all of which experience vast sums of fraudulent purchases that cost the companies millions in lost revenue each year. Comsec were tasked with the prevention of these losses, so a move to AWS that could provide tools and features to help in targeting these fraudsters was a necessity.

What our customers say

Hear directly from those who’ve experienced our services. Discover how we’ve made a difference for our clients.

Solution

Creating the AWS shell

As an initial setup, this project took on the shape of forming the core infrastructure and skeleton to which AWS services could be deployed and developed. The infrastructure brought with it AWS Lambda serverless compute services, AWS Step Functions, AWS key Management Service (KMS) and Amazon S3 buckets (to name a few).

The infrastructure and backend were made deployable via CICD (i.e GitHub Actions) and all processes were heavily documented to assist with ongoing learning for the Comsec team. This skeleton will quickly become fleshed out with the integration of machine learning services (Amazon Translate, Amazon Augmented AI (A2I) and Amazon Comprehend, to be precise). As well as helping to power vital data insights through visual database services like Amazon Neptune.

Comsec were looking for quality at speed for this project, and our team were able to deliver well within the estimated timeframe. We were able to provide not only the core structure which will power their future fraudulent targeting and machine learning modelling, but also provide rapid training on AWS turnkey solutions.

Knowledge is power

Equally important to the AWS infrastructure setup was the training and documentation for the handover. The Firemind team provided comprehensive documentation to the Comsec team, covering new services and machine learning tools. This ensures Comsec can confidently understand and build on the infrastructure, saving time and costs in future deployments.

A guide to frameworks

Similar to the first key value add, framework training will prove to be vital to the Comsec team as they venture into more serverless workflows and machine learning modelling. This project was Comsec’s first taste of many framework standards, such as Terraform and GitHub Actions, allowing them to build more automated processes within the AWS cloud for future projects.

Entry to managed services

This project has seen Comsec begin to harness the power of managed machine learning services, such as Amazon Comprehend (for natural language and sentiment processing) and Amazon Translate (for highly customisable language translation).

Get in touch

Want to learn more?

Seen a specific case study or insight and want to learn more? Or thinking about your next project? Drop us a message!