unthinkable
Case study · A healthcare provider, London, UK · personalised care app for cancer survivors

Cancer survivors using the app get updates and fixes sooner, with less downtime.

The provider wanted cancer survivors to keep getting care from its dieticians, physiotherapists, psychologists and nurses through an app that improves quickly and stays up.

London, United Kingdom

What changed

20%increase in deployment frequency (as reported)
5%change failure rate (as reported)
30 minsaverage time to recover (as reported)
What stood in the way

What stood in the way.

What stood in the way
  • Each release meant typing the branch name into AWS CodePipeline and clearing the CloudFront cache by hand, which caused downtime and production issues.
  • Too little monitoring data slowed root cause analysis, and infrastructure was provisioned by hand.
What changed
  • 20%increase in deployment frequency (as reported)
  • 5%change failure rate (as reported)
  • 30 minsaverage time to recover (as reported)
What we built

What we built.

  • A Lambda function, triggered by CloudWatch events when a branch is created in CodeCommit, that deploys a CloudFormation stack with a new pipeline for that release.
  • Metrics, dashboards, log management and alarms in CloudWatch, with activity tracking in CloudTrail.
  • Secrets in AWS Secrets Manager and encryption with AWS KMS, with workloads tuned to keep costs down.
  • AWS Lambda
  • AWS CodePipeline
  • AWS CodeCommit
  • AWS CloudFormation
  • Amazon CloudWatch
  • AWS CloudTrail
  • Amazon CloudFront
  • AWS KMS
What it shows

What this work shows about us.

Release speed is patient care

When the app is a care channel, downtime is missed support. Automated pipelines removed the manual steps that caused it.

Evidence when things go wrong

Monitoring and activity tracking cut the time to find a cause to a 30-minute average recovery (as reported).

Want to automate your app release cycle on AWS Lambda? Tell us what you want to change. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with the right team.

Start your plan ↗
If we built it today

If we built it today, with AI.

This work came before today's AI. If the London-based healthcare provider for cancer survivors asked us now, this is what we would add, step by step.

Emergency

An agent would read the alarms and logs when a release fails, find the change that caused it and prepare the rollback for the engineer to approve.

Going home

An agent would tell patients and care staff when a fix they reported has been released. Engineers still decide what ships.

Ask Unthinkable

Want this for your business?

Tell us what you run or what you're building. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with a senior AI engineer.

Send us a note instead

Required

Thanks. A senior AI engineer will reply to you.
u
Ask UnthinkableClarity first, then the right team
Your conversation

An AI agent. This prototype uses prepared answers and sends nothing.