AI security engineer / production systems

andres garcia

I build AI systems that survive production: evals that block bad merges, guardrails, and the observability to prove it.

/prompt injection, regressions, cost, drift/

A CI merge gate that re-runs a golden dataset on every prompt or model change and blocks the merge on a critical regression.

02 / Latest writingView all 1 →

A one-paragraph prompt edit looked harmless and broke 9 test cases. How I built a CI gate that treats prompts like code and blocks the merge.

03 / Working stack6 tools

01Python

02Anthropic API

03GitHub Actions

04Docker

05SQLite

06Astro

Open to consulting on AI production systems.

let’s talk