Skip to content

AWS

When the audit selects AWS, we build the infrastructure layer from these native services — all codified with Terraform/OpenTofu modules.

  • AWS Organizations with separate accounts to isolate environments and blast radius (e.g. production vs. non-production, plus a shared/management account).
  • Guardrails and baseline policies applied org-wide.
  • VPC with public/private subnets across multiple Availability Zones.
  • Route tables, internet and NAT gateways, and security groups.
  • Private networking for data stores so they’re never directly exposed.
  • IAM roles and policies, least-privilege by default.
  • IRSA (IAM Roles for Service Accounts) so Kubernetes workloads assume scoped AWS permissions without long-lived keys.
  • Amazon EKS as the managed control plane for the containerization layer.
  • Managed node groups and/or Fargate, with cluster autoscaling.
  • RDS / Aurora for relational databases — multi-AZ, automated backups, point-in-time recovery.
  • S3 for object storage, artifacts, and backups, with lifecycle policies.
  • ALB / NLB for load balancing, fronting the cluster ingress.
  • Route 53 for DNS, ACM for managed TLS certificates.
  • KMS for encryption keys; encryption at rest enabled across data stores and volumes.
  • Secrets Manager for application and platform secrets (surfaced into the cluster via the secrets workflow).
  • CloudWatch for cloud-level metrics and logs at the account boundary — complemented by the full Grafana observability stack running in-cluster.