How we work One engineer, a written plan, and no black box.
Every engagement runs the same four steps and ends with something you could hand to another firm.
The shape of it
Four steps, every time.
No discovery phase that turns into an invoice, and no hand-off to a delivery team you have never met.
-
Scope
A thirty-minute call, a written scope, and a fixed price. If the audit is the wrong first step, we say so on this call.
-
Access
Read-only by default: a cloud role that can list and read, cluster read, repository read, and a billing export. Write access is agreed in writing and limited to the work.
-
The work
One engineer, doing the work, with a written checkpoint halfway. No status meeting that exists to explain why nothing moved.
-
Handover
A walkthrough, the documents, and the credentials. You keep everything, including the ability to take it elsewhere.
The deliverable
What you keep when the work is done.
The difference between a consultant and a contractor is what is left behind. Ours is written down, in your repository, in your name.
- The report. Findings ranked by dollars, risk, and effort, with the evidence behind each one.
- The SLOs. Written down, agreed with product, and wired into your monitoring.
- The code. Terraform and Kubernetes manifests in your repository, not ours.
- The runbooks. One for every alert that actually pages someone, written for the person who gets the page.
- The evidence. Logs, exports, and records an auditor accepts, produced by the platform on a schedule.
Boundaries
What we don't do.
Some of these are the reason to hire us. All of them are the reason you can trust the rest.
We don't sell 24/7 on-call
We design the rotation, the escalation path, and the runbooks, and we are available in business hours. Round-the-clock human staffing is a team, and one consultant with a day job cannot be one.
We don't lead with migrations
A six-week migration is a poor first engagement for a company that has not measured its platform yet. Audit first, then the work the audit justifies.
We don't install an appliance
No proprietary box, no per-seat licence, no dashboard that stops working when you leave. The tooling is open and the configuration is yours.
We don't promise a smaller bill
We find the spend that is not buying anything and put a number on it. If the audit finds nothing worth acting on, the report says that too.
Tooling
What we leave behind.
Everything here runs in your accounts, under your control, and stays when we go.
- Orchestration
- Kubernetes, AKS, EKS, Docker
- Infrastructure as code
- Terraform, Ansible, ARM templates
- Delivery
- Argo CD, GitHub Actions, Azure DevOps, Helm
- Observability
- Prometheus, Grafana, Loki, Azure Monitor, CloudWatch
- Compliance
- OPA, Vault, audit logging, evidence collection
- Networking
- VPC and VNet design, ExpressRoute, BGP, service mesh, zero trust access
Start with the audit?
One week, fixed price, read-only. If it is the wrong first step for your platform, we will tell you before you buy it.