Engagements

Fixed scope, fixed fee, read-only access.

Four kinds of engagement across AWS and Azure. Each one ends in something your team can act on — a ranked backlog, a risk matrix, a working stack — not a slide deck of recommendations.

Book a discovery call Hiring instead?

01 — 5–7 business days

FinOps & Cloud Cost Audit

Find what your AWS or Azure estate is actually costing you, and what to do about it.

AWS Cost Optimization Audit A full inventory sweep of your AWS accounts, ending in a ranked remediation plan. Azure Cost Optimization Audit The same sweep across Azure subscriptions, with Azure’s own cost mechanics in mind. Data Transfer & NAT Gateway Cost Reduction The cost line nobody owns, and the one most often accidental. Database & RDS Cost Optimization Databases are usually the largest single line, and the least examined. Kubernetes & EKS Cost Optimization Cluster spend is a bin-packing problem before it is a pricing problem. Savings Plans & Reserved Instance Strategy Modelling the commitment decision properly, before you lock in three years.

02 — 7–10 business days

Well-Architected & DevSecOps Review

Security posture, resilience and pipeline integrity, ending in a risk matrix you can action.

AWS Well-Architected Review The six pillars, run properly, with findings graded and costed. CI/CD Pipeline Security Review The pipeline has production credentials. It deserves the same scrutiny as production. Container & EKS Security Review Cluster security is mostly configuration, and mostly default. IAM & Least-Privilege Review Permissions accumulate. This is the audit that removes what is no longer needed. SOC 2 Readiness on AWS The infrastructure half of SOC 2, handled before the auditor arrives.

03 — 1–2 week sprint

Production Observability Setup

Know something is wrong before your customers tell you.

Observability Cost Reduction Datadog, New Relic or CloudWatch bills that have outgrown their value. OpenTelemetry Instrumentation Vendor-neutral tracing, so the backend decision stays reversible. Prometheus & Grafana Implementation A self-hosted stack that stays affordable as cardinality grows. SLO & Alerting Design Fewer alerts, each of which means something.

04 — Monthly, 2-month minimum

Fractional Cloud Architect

Principal-level architectural judgement on retainer, without the headcount.

Post-Incident Architecture Review After the outage, the harder question: what made it possible? Scaling Architecture Advisory Ongoing design review for teams growing faster than they can hire. Technical Due Diligence An independent read on an architecture before money moves.

— — Also available

Hiring cloud engineers rather than fixing infrastructure?

I also run technical interview rounds for organisations hiring cloud and DevOps engineers — a scored written recommendation per candidate, with no placement fee and no interest in the outcome.

See how interviewing works →