About

I build production infrastructure for AI agents — the IAM boundaries, deployment pipelines, and guardrails that decide whether an agent system survives contact with real traffic. This site is where I write up the patterns as I ship them.

What I write about

Agent security & IAM

Scoping agent credentials to the exact permissions they need, and the failure modes that show up when you don’t — prompt injection, runaway tool calls, secrets exposure, and the audit trail you need when something goes wrong.

Infrastructure as code for agent systems

Terraform patterns for agent infrastructure that survive real change velocity: modular workspaces, state management, and environments that don’t need a human to babysit them.

CI/CD for agent deployments

Evaluation gates, staged rollouts, and rollback triggers — deployment pipelines built for systems that can make their own decisions, so confidence is earned instead of assumed.

Background

Full-stack engineering background, now focused on the infrastructure and security layer under production AI agents. I also do consulting and engineering-leadership work through Anonymous Function Labs, where the same patterns get applied at client scale.

Get in touch

Reach out via the contact page or any of the channels in the sidebar. If you’d rather start with something concrete, the patterns from this site are packaged as skills on the skills hub.