Cloud and DevOps on AWS and Azure
Reproducible infrastructure, deployable daily, with the bill under control.
At Pulselive I automated infrastructure with Terraform on AWS; at BASF and Capitole I took containerised systems to Kubernetes and OpenShift. The difference between a cloud that helps and one that bleeds money is almost never the technology: it is whether the infrastructure is defined in code and whether anyone looks at the bill.
I work from least privilege by default. I have seen enough deployments with admin permissions handed out because it was quick, and enough bills inflated by resources nobody switched off, to treat it as anything but a detail.
What's included
Infrastructure as code
Everything defined in Terraform, versioned and reproducible. Standing up an identical environment stops being a project and becomes a command.
Containers and orchestration
Docker and Kubernetes (or OpenShift) with lean images, resource limits and zero-downtime deployments.
CI/CD pipelines
GitHub Actions, GitLab CI or Jenkins: tests, analysis, build and automatic deployment. Deploying stops being an event.
Security and permissions
Least privilege in IAM, secrets out of the code and into a secret manager, and a review of what is exposed to the internet.
Observability
Structured logs, metrics and alerts that reach you before your customer does.
Cost optimisation
Bill review, right-sizing, lifecycle policies and switching off what nobody uses.
How we work
- 01
Initial audit
I review what you have: architecture, permissions, pipelines and cost. You get a report with findings prioritised by impact.
- 02
Work plan
What gets fixed first, what can wait and what is not worth touching. With effort and price per block.
- 03
Execution
Incremental, reversible changes. No all-or-nothing migrations on a Friday afternoon.
- 04
Handover
Infrastructure documentation and training so your team can operate it without me.
Stack
- AWS
- Azure
- Terraform
- Docker
- Kubernetes
- OpenShift
- GitHub Actions
- GitLab CI
- Jenkins
- S3
- Lambda
- SQS
- CloudFront
- IAM
Frequently asked questions
Can you reduce our AWS bill?
In most cases yes, and it usually comes from the same places: oversized instances, storage with no lifecycle policies, test environments left running overnight and badly planned data transfer. The initial audit tells you how much is on the table before you commit to anything else.
AWS or Azure?
I have worked deeply with both. If you are already on one, the technical differences almost never justify migrating. If you are starting from zero, the decision usually depends more on your team and your commercial agreements than on the technology.
Do we need Kubernetes?
Only if you have enough services and enough team to justify its operational cost. For many projects, containers on a simpler managed service do the same job with far less maintenance.
Can you work with our infrastructure team?
Yes, that is the normal case. I come from distributed international teams and I am used to fitting into processes that already exist rather than imposing my own.
Is your infrastructure costing more than it should?
Tell me how it is set up and I will tell you where the problems are. First call is free.