Zero-trust cloud, in four to six weeks.
A fixed-scope engagement that builds the foundation a serious South African enterprise actually needs — AWS af-south-1 architected for POPIA Section 72 from the VPC up, Terraform infrastructure-as-code your team owns, GitHub Actions CI/CD that ships safely, and the observability stack to know when something is wrong before customers do. Built on the same patterns sonofgraig uses to run its own platform.
What ships at the end of week six.
Most cloud-architecture engagements fail in the same three ways: the deliverable is a slide deck instead of running infrastructure, the security posture is documented but not enforced, and the operations team gets a system they cannot actually run. We solve all three by making them deliverables — not optional add-ons.
Cloud Architecture & DevOps Solutions, in one engagement.
sonofgraig's source-service catalog defines Cloud Architecture and DevOps Solutions as twelve sub-disciplines. The standard scope below covers the six that matter most for an enterprise establishing or rebuilding their cloud foundation. The other six are quoted as add-ons during scoping.
- VPC, subnets, NAT, ALB, security groups
- Multi-AZ deployment for production
- Region-lock IAM SCP — no cross-region provisioning
- EKS cluster with managed node groups
- Helm charts for app deployment & rollback
- Cluster autoscaler & HPA configured
- Terraform modules per environment
- Remote state with locking & encryption
- Plan output enforced in PR before apply
- Build · test · scan · deploy in pipeline
- Helm chart deploys with rollback
- Branch protection on protected branches
- Four golden signals dashboarded
- SLO & error-budget tracker per service
- Sentry + OTel for tracing & errors
- Cloudflare WAF, DDoS, rate limiting
- OIDC federation — no long-lived keys
- Secrets Manager + KMS for all credentials
Five layers, all defined in code.
Every layer below is delivered as Terraform modules in your repository. Each layer has a defined purpose, a defined boundary, and a defined responsibility for either security or reliability. You inherit it as code — nothing is delivered as a black box or an undocumented manual configuration.
Three regions, all in South Africa.
sonofgraig builds on AWS af-south-1 by default because that's the region sonofgraig's own platform runs in. If your group security policy mandates Microsoft or Google, we deliver the equivalent architecture on Azure South Africa North or GCP Johannesburg — the same outcomes, the same scope, the same fixed price.
- EKS, RDS, S3, Secrets Manager, KMS, ALB
- Terraform AWS provider — mature ecosystem
- SCP-based region lock to af-south-1
- Cost Explorer & Budgets pre-configured
- OIDC federation from GitHub Actions
- sonofgraig has shipped this 30+ times
- AKS, Azure SQL, Blob Storage, Key Vault
- Terraform AzureRM provider — equivalent modules
- Region restriction via Azure Policy
- Cost Management + budgets
- Workload identity federation from GHA
- Add 5 business days to delivery window
- GKE, Cloud SQL, Cloud Storage, Secret Manager
- Terraform Google provider — equivalent modules
- Org Policy constraints lock to africa-south1
- Billing budgets & quotas
- Workload identity federation from GHA
- Add 5 business days to delivery window
Multi-cloud (resources in two clouds simultaneously) is out of standard scope but quotable as an extension. sonofgraig's Cloud Console product (Q2 2027) provides multi-cloud unified management as a subscription.
Reviewable. Reversible. Yours.
A real Terraform module from a delivered engagement, slightly redacted. Your team gets the full module library at handover — networking, EKS, RDS, IAM, observability, CI — with documentation and a runbook explaining how to extend it. There is no proprietary wrapper to learn.
dev, staging, prod with isolated state# sonofgraig Cloud Architecture & DevOps engagement # Network module — production VPC for af-south-1 # Terraform >= 1.6 · AWS provider >= 5.0 terraform { required_version = ">= 1.6" required_providers { aws = { source = "hashicorp/aws", version = "~> 5.0" } } } module "vpc" { source = "terraform-aws-modules/vpc/aws" version = "~> 5.5" name = "${var.org}-${var.env}-vpc" cidr = "10.0.0.0/16" azs = ["af-south-1a", "af-south-1b", "af-south-1c"] private_subnets = ["10.0.1.0/24", "10.0.2.0/24", "10.0.3.0/24"] public_subnets = ["10.0.101.0/24", "10.0.102.0/24", "10.0.103.0/24"] database_subnets = ["10.0.21.0/24", "10.0.22.0/24", "10.0.23.0/24"] enable_nat_gateway = true single_nat_gateway = false # multi-AZ NAT — production enable_vpn_gateway = false # POPIA s.72 — flow logs to detect cross-region egress enable_flow_log = true flow_log_destination_type = "s3" flow_log_destination_arn = aws_s3_bucket.flow_logs.arn flow_log_traffic_type = "ALL" flow_log_log_format = null tags = var.tags } # Region-lock SCP — applied at the org level via separate # identity module. Prevents accidental provisioning in # any region except af-south-1. Override requires explicit # approval through the security team's IAM role.
Four phases. Four to six weeks.
Every phase ships a tangible deliverable. The cadence below is the standard plan; complex compliance regimes (financial services, healthcare) or non-AWS clouds may extend the build phase by up to a week. We will tell you which week you are in at every check-in.
terraform plan on every pull request. Region-lock SCP applied at the AWS organisation level.Exactly what's in. Exactly what's not.
Fixed-scope means we have to be explicit about boundaries. The lists below are the standard inclusions and exclusions for the R75,000 starting price. Anything in the right column can be quoted as a separate engagement — or rolled into a Cloud Console / Pipeline Builder subscription on conversion.
- Discovery workshop, scoping document, POPIA risk register
- Single AWS account in af-south-1 (or Azure SA North / GCP africa-south1)
- Production VPC with multi-AZ private/public subnet topology
- Region-lock SCP at the AWS organisation level
- Identity Center / SSO integration with your IdP
- EKS cluster with managed node groups & cluster autoscaler
- Single RDS PostgreSQL instance with encrypted snapshots
- S3 buckets with versioning, lifecycle policies, KMS encryption
- Cloudflare WAF, DDoS, and rate limiting at the edge
- OIDC federation from GitHub Actions to AWS — no static keys
- End-to-end CI/CD for one reference workload
- SAST (semgrep, gitleaks) & SCA (trivy) gated in CI
- Prometheus, Grafana, OpenTelemetry, Sentry — configured
- Four-golden-signals dashboard & SLO tracker
- CIS AWS Foundations assessment with remediation
- Runbooks: incident response, rollback, on-call rotation
- POPIA Section 72 evidence pack
- Two knowledge-transfer sessions with your team
- 30 days of priority support from go-live
- Application development — we deploy yours, we don't write it
- Migration of existing workloads — quoted per workload
- Multi-cloud (resources spanning AWS + Azure or AWS + GCP)
- Cybersecurity penetration testing by a third-party assessor
- SOC 2 / ISO 27001 audit preparation work beyond foundations
- 24/7 SOC monitoring — available as a Cluster 2 retainer
- Disaster recovery drills beyond initial runbook validation
- On-premises or hybrid network engineering (VPN, Direct Connect)
- AWS / Azure / GCP cloud spend — billed to your account
- Third-party SaaS subscriptions (Datadog, New Relic, etc.)
- Long-running operational support beyond the 30-day window
DORA metrics, measured at handover.
The four DORA metrics are the industry-standard view of high-performing engineering organisations. Below is what a typical engagement delivers as a starting baseline — not aspirational targets, but what the architecture actually achieves on the reference workload at go-live. Your team builds from this baseline, not from zero.
main. Rate-limited only by the team's velocity, not by infrastructure.Values shown are typical baseline performance for the reference workload at go-live. Real-world numbers depend on your team's testing discipline, release cadence, and incident response practices. The architecture creates the conditions for elite performance — the team has to use them.
Compliance designed in — not bolted on later.
South African enterprise procurement teams ask two questions about any cloud architecture: where does the data live, and who can prove it? The four cards below are the controls every engagement ships with, and the evidence pack you receive at handover.
Independent reference: sonofgraig publishes a complete POPIA compliance statement and a long-form security overview. The architecture in this engagement matches both documents — not a simplified version of either.
An opinionated stack. Open source where it matters.
We do not invent the engine; we invest where the value is. Every component below is production-grade open source or a managed service we consciously chose not to rebuild. You inherit the same engineering decisions sonofgraig's own platform was built on — and you keep the source code.
One number. No hourly surprises.
sonofgraig service projects are deliberately simple to procure. The price is the price. Scope is fixed before contracting. Variations are quoted in writing and signed before any additional work is performed.
Questions procurement, security and platform teams ask.
If your team is preparing for a vendor review or a board sign-off, the answers below cover most of what gets raised. Anything else, your account team can route to engineering directly.
Is the R75,000 fixed, or just a starting figure?
Do you charge for the AWS / Azure / GCP cloud spend?
Why AWS af-south-1 and not Azure or GCP by default?
Who owns the Terraform code at handover?
Can the engagement be extended if scope changes?
What does the 30-day post-delivery support cover?
Do you provide a SOC 2 / ISO 27001 audit dossier as part of this?
Can you migrate workloads from our existing cloud or on-premises?
What identity providers do you integrate with?
Are sonofgraig B-BBEE certified and CIPC registered?
Book a 30-minute scoping call.
A senior solutions engineer joins, we step through the workloads you intend to run and your compliance constraints, identify whether it fits the standard scope, and confirm what your final fixed price will be. No commitment until contract signature.