Sergio Rodriguez

What it shows: A support queue that distinguishes a reported fix from current evidence and a matching resolution receipt.

My work: Interactive synthetic requests, decision rules, reproducible checks, and a runbook with a next action for stale records or mismatched receipts.

Scope: A demonstration using synthetic records and a simulated clock.

Inspect: sergrdz.pages.dev/demos/support-intake/ — demo, checks, fixtures & runbook

What it shows: A decision gate that checks whether a proposed action matches a signed approval, including its caller, target, arguments, expiry, and prior use.

My work: A Python reference implementation, a persistent approval ledger, controlled fixtures, and tests for validation and replay behavior.

Scope: A local security experiment. It returns a decision without executing the proposed action.

Inspect: github.com/SSH-PuR66/countersign — source, tests & reproduction

What it shows: How source code, decompiler output, and observed native behavior can differ—and how to investigate the discrepancy.

My work: An original C fixture, native tests, a Ghidra export script, and a comparison of recovered types and boundary behavior across two compiler builds.

Scope: Controlled original fixtures. Browser inputs use an arithmetic model; recorded native results are separate.

Inspect: github.com/SSH-PuR66/cutline — source, native results & method