AI security & systems engineering · Founder, CryptoFixYup
I build software that puts explicit controls around AI-generated decisions. My focus is deterministic authorization, lifecycle state machines, cryptographic integrity, and reproducible engineering evidence.
Rust · TypeScript / Node.js · Python · GitHub Actions
| Project | What it contains |
|---|---|
| Alpha Sentinel | AI-agent authorization research: policy checks, transaction-bound approvals, simulation boundaries, lifecycle handling, and regression tests. |
| Autonomous Operations System | Logistics-focused scaffold with deterministic business constraints, an AI orchestration boundary, and operational APIs. |
| market-monitor-rs | Rust market-signal components with validation for invalid prices and a CI workflow for formatting, compilation, linting, and tests. |
| pelican-rs | Rust telemetry wire-decoder components with length, version, checksum, and semantic validation. |
- Keep model reasoning separate from execution authority.
- Make state transitions and failure modes explicit.
- Turn discovered defects into regression tests.
- Tie results to the exact code, fixtures, and environment evaluated.
- Distinguish prototypes, passing tests, and demonstrated production behavior.
Alpha Sentinel is under active development. Its README documents implemented components, historical CI evidence, and integration work still to be demonstrated.
Repository security reviews, CI repair, and bounded reliability improvements for small engineering teams.



