Litigation paralegal and independent researcher working at the intersection of AI systems, legal evidence, and software development. I build and contribute to tools for source-grounded legal research, agent reliability, provenance, and auditable workflows.
My interests include AI and agentic-system discovery, preserving execution evidence, reproducible evaluations, legal research infrastructure, and responsible use of AI in legal practice.
Discovering Agentic AI: Discovery, Preservation, and Proof in AI and Agentic Systems
Independent white paper · Law Archive (OSF Preprints) · 2026
Examines how evidence from AI and agentic systems can be identified, preserved, reconstructed, and evaluated in the context of U.S. civil discovery and evidentiary practice.
- Implicit Core — Python infrastructure for addressable, versioned AI-agent experiences, selective state materialization, provenance, and reproducible evaluation. Published as
implicit-ai. - Pro-One — Development-stage, source-grounded legal research and procedural intelligence, with versioned evidence, citation checks, and human-review boundaries. Not a live legal advice service.
- Free Law Project — Merged contributions to Litigant Portal and CourtListener:
- Harvey LAB — Merged a runtime dependency fix for model-adapter packaging; submitted an evaluation-reporting fix to generate scoped comparisons and surface failures (open for review).

