AI Safety Research Engineer · Cognitive Governance Architect · Founder, Purple Brain Labs
I build governance systems for AI work that must remain inspectable, contestable, and truthful under real-world pressure.
My current work centers on The Living Constitution 2.0: a governance control plane for organizing AI safety research, C-RSP contracts, evidence, product modules, reusable research templates, and public reviewer surfaces.
| Path | Use This For | Link |
|---|---|---|
| Mothership | Governance system, registry, contracts, templates, truth state | The Living Constitution 2.0 |
| Public Site | Public-facing Living Constitution portfolio and Folio 001 | coreyalejandro.com |
| Module Status | What is working, partial, draft, planned, or unverified | MODULE_STATUS.md |
| Static Site Repo | Public static Living Constitution site source | the-living-constitution-2.0-portfolio |
I work at the intersection of:
- AI safety research
- runtime governance
- evidence systems
- human-AI interaction
- cognitive safety
- research-engineering infrastructure
- accessibility-aware tool design
- portfolio systems that distinguish verified work from drafts and prototypes
The Living Constitution 2.0 is the mothership of my current work.
It governs:
- C-RSP contracts
- module registries
- artifact registries
- route registries
- evidence records
- verification outputs
- research-to-paper-to-product templates
- public/private boundaries
- portfolio-safe public data
C-RSP means Constitutionally-Regulated Single Pass. It is one subsystem inside The Living Constitution 2.0, not the whole system.
| Surface | Status |
|---|---|
| The Living Constitution 2.0 repo | Working local governance/control-plane scope |
| Public site at coreyalejandro.com | Working static public surface |
| Folio 001 | Static prototype / origin exhibit |
| Research-to-paper-to-product template | Partial, usable template scaffold |
| HIDRS study repo | Active research module, partial |
I do not claim that every registered module is production-ready.
I do not claim that every research project is complete.
I do not claim that every public surface is final.
I do not expose private Lab mechanics, raw incident transcripts, unpublished drafts, secrets, or unreviewed evidence chains as public claims.
Research reviewers should start with The Living Constitution 2.0 and the system diagrams in its README.
Build reviewers should inspect the module registry, templates, verification scripts, and public site repo.
Evidence reviewers should start with `MODULE_STATUS.md`, registry files, and public-safe artifacts.
Portfolio reviewers should start at coreyalejandro.com.
I am building a governed AI safety research-and-product system where claims, evidence, prototypes, working modules, and public-facing surfaces are separated by explicit truth status rather than collapsed into one polished but misleading portfolio.