The platform for the care sector, built on AWS

A Healthcare & Life Sciences business working with Cloud Combinator on AWS. The client is anonymised at their request.

The production engagement had three focus areas.

Cloud Combinator retained the validated core from the proof of concept and wrapped it in the network, security, resilience and observability layers a production system needs, all deployed repeatably through AWS CloudFormation across development, staging and production.

The proof of concept was completed and accepted, demonstrating the AI reasoning approach in practice. The production phase is scoped against a clear set of acceptance thresholds, shown below as targets, that define what the hardened system must achieve before sign-off.

• Explainable by design — Every recommendation carries clear reasoning, an alternative and the individual home analyses as an audit trail, with the professional making the final call.
• Risk-aware — The advisor is built to catch risk interactions and peer dynamics that statistical matching alone would miss, the specific value the concept proved.
• Secure and UK-resident — Encryption at rest and in transit, least-privilege IAM, Cognito authentication, WAF, KMS and CloudTrail, with data kept in the eu-west-1 and eu-west-2 regions and none retained between invocations.
• Resilient under load — SQS with a dead-letter queue, retries with exponential backoff and a circuit breaker on Bedrock, targeting 50 concurrent requests at under an one percent error rate.
• Handover — Repeatable CloudFormation deployments, dashboards and alarms, plus runbooks and knowledge-transfer sessions so the client can operate the system independently.

With a validated concept now being hardened for daily use across the estate, the client has an AI advisor that is not just a demonstration but a production system, ready to support care professionals in one of their most consequential decisions.