I'm a security leader working on product security, GRC, and how to let AI coding agents do real work without giving up control. Off the keyboard, I'm a private pilot in the Pacific Northwest.
What I work on
- Guardrails for running AI coding agents with fewer prompts and the same safety
- Making AI-assisted code traceable: who, or what, wrote a change
- SDLC controls that produce audit evidence as a side effect, not a scramble
- Open-source GRC tooling
Projects
- safe-yolo: local guardrails for running Claude Code and Codex yolo-style, with reversible snapshots, a denylist hook, and OS sandboxing
- ai-attribution-hooks: drop-in git hooks that make AI-assisted code attribution queryable
- sdlc-control-gates: a catalog of SDLC control gates and the evidence each one emits, mapped to SOC 2, ISO 27001, PCI DSS 4.0, and HIPAA
- security-atlas: an open-source GRC platform
Writing
- mattgoodrich.com: field notes on AppSec, workload identity, AI governance, and shipping software with coding agents
- productsecurity.dev: articles for people learning and working in product security
- wingsoverpnw.com: flying and aerial video from the Pacific Northwest



