AI-first engineering with Claude Code
I use Claude Code every day. I'm a DevOps and cloud engineer who also builds and ships working applications with it, so you get complete tools, not only the infrastructure under them.
What it looks like in practice
- Writing and refactoring GitHub Actions and Azure DevOps pipelines.
- Terraform and OpenTofu changes, reviewed and planned before any apply.
- Automating repetitive GitHub work, from repository setup to pull request housekeeping.
- Python, PowerShell and Bash scripts for operations work.
- Small internal tools and applications, built end to end.
- Working inside development squads as the person who connects code, pipeline and cloud.
The guardrails
Speed only helps if the result is safe to run. These rules apply on every engagement:
- Every change gets human review before it merges.
- Secrets never go into a prompt.
- Veracode and SonarQube gates apply to AI-written code exactly as they apply to any other code.
- Infrastructure changes need human approval before they are applied.
- Your company's AI policy comes first. If a tool or a use is not allowed, I don't use it.
Who this is for
Teams on Azure that want to use AI in delivery but need controls they can show to security, compliance or an auditor. That is common in fintech, payments and healthcare, where I have spent most of my career.
Let's talk
Want to try AI in your delivery without losing control of it?