Infrastructure that could keep up with the growth
Kubernetes, secrets management and observability reworked for one of the fastest-growing HRMS platforms in the USA.
01 — The challenge
Where it started.
The platform was growing quickly and scaling by adding instances, because that was the lever available. The Kubernetes estate had grown with it rather than being designed for it.
It also holds sensitive employee data, and secrets and logging were handled differently in different environments — the condition under which an incident becomes a disclosure.
02 — What we built
The work itself.
Every environment on EKS Fargate. Managed, node-less compute so capacity follows demand rather than being provisioned ahead of it.
Helm charts as the unit of release. The same chart promoted between environments, so what runs in production is what was tested.
HashiCorp Vault. Credentials centralised and rotated rather than distributed through configuration, across every environment.
Centralised logging, monitoring and alerting. One place to look during an incident, with MongoDB and PostgreSQL workloads covered alongside the application.
03 — Results
What the client has now.
High availability, with scaling that tracks demand instead of being provisioned for peak. Consistent secret handling across environments, with employee data held to one standard. Monitoring and alerting covering the platform rather than its newest components.
Looking for a similar solution? Contact us today!