← Work

2024 · Infrastructure

Parallel production and staging infrastructure on AWS

a consumer web platform · Devops engineer

The problem

Staging had to be a real rehearsal of production, with no shared network or cache that a staging mistake could reach production through.

What I built

I built two fully separate environments from the same code, each with its own network, container service and cache, with private access for engineers.

The result

The team can test risky changes in staging with no shared network or cache with production.

Main tools

  • Terraform
  • Amazon ECS
  • Load balancer
  • ElastiCache
  • AWS Client VPN

This is the short version. The full case study is available on request.

Ask for the full case study →