atomsized

Selected AWS platform work.

Case studies / Selected work

Case studies showing the architecture, implementation decisions, and operational outcomes.

Schedule a platform callPreview / UAT
Hand-drawn project sheets with architecture, delivery evidence, decisions, and handover notes

Selected case files

Case file 01

SaaS platform

Cost-aware SaaS platform

Outcome

A reviewable path from low-cost stage to production-ready AWS boundaries, owned in code.

Starting point
Stage needed to stay economical without turning production into a separate platform design.
What we built
Separate stage and production foundations, Argo CD app-of-apps, Karpenter, managed secrets, and environment-specific data services.
Our role
AWS architecture, Terraform foundations, Amazon EKS, and GitOps delivery design.
TerraformAmazon EKSArgo CDKarpenterAWS
Read the case study
Hand-drawn AWS GitOps architecture connecting a cost-aware stage environment to production-ready services
Case file 02

Major fast-food brand

Serverless customer-contact workflow

Outcome

A lean contact workflow with private attachments, validation, and visible failure signals.

Starting point
A public contact flow needed secure image uploads and reliable routing without an always-on application.
What we built
A protected serverless path from CloudFront and WAF to Lambda, private S3, email, and alerts.
Our role
AWS architecture, implementation, infrastructure as code, security, and observability.
CloudFrontAPI GatewayLambdaS3AWS WAF
Read the case study
Hand-drawn serverless contact workflow with edge protection, API, Lambda, private storage, email and alerts
Case file 03

SaaS platform

Microservice preview automation

Outcome

Teams can launch one-service previews or coordinated UAT stacks from a pull request.

Starting point
Cross-service changes needed realistic review environments without repeated manual setup.
What we built
Pull-request labels, an automation repository, Argo CD, Helm, DNS, data services, and automatic teardown.
Our role
Control-plane design, ApplicationSets, lifecycle automation, and handover.
Argo CDApplicationSetsHelmGitHub ActionsKubernetes
Read the case study
Hand-drawn GitOps delivery loop with Preview and UAT environments before production

Common engagement patterns

Start from the closest platform constraint.

Each pattern is adapted to the accounts, repositories, risk, and ownership model already in place.

AWS foundations and modernisation

Accounts, identity, networking, runtimes, Terraform, backups, and budgets.

GitOps and AI delivery workflows

Evidence, pull requests, CI, Helm, Argo CD, verification, and rollback paths.

Reliability signals

CloudWatch, Prometheus, Grafana, alerts, SLOs, and runbooks.

Security and cost

Public controls, identity review, security reports, distributed load tests, budgets, and remediation notes.

After handover

The architecture, automation, and operating context stay with your team.

Use these case studies to compare the closest delivery pattern with your AWS accounts, repositories, constraints, and ownership model.

Schedule a platform call