I build document-first systems for AI-assisted work where authority, evidence, review, provenance, and human approval need to remain visible.
My strongest work is not model training or autonomous-agent development. It is the operating layer around AI: workflow design, human gates, documentation architecture, quality control, behavioural continuity, claims discipline, and recovery of useful systems from messy source material.
A five-paper public proof-of-work collection covering behavioural continuity, controlled art archives, human-controlled small-business infrastructure, continuity and handoff, and conversational work translation.
A twelve-role public reference architecture with structured handoffs, gate levels, tests, provenance controls, review logic, and explicit human authority.
A bounded, tested Python prototype built around visible symbolic mutation, deterministic test modes, and deliberately narrow technical claims.
The public index for selected PPLL systems, art and licensing information, and approved releases.
- human-gated AI workflows
- workflow governance and approval design
- documentation and knowledge architecture
- AI claims auditing and evaluation
- behavioural continuity and external operating controls
- provenance, verification, and quality-control systems
- operational and plain-language writing
- capability recovery from large or messy archives
- unusual contained prototypes
Public repositories contain selected proof of work. Personal records, credentials, private source archives, commercial records, unreleased systems, and non-public implementation material stay outside the public GitHub.
My operating approach is informed by more than twenty years in food service and hospitality, along with formal study in Culinary Management, Business Management and Entrepreneurship, and Hotel and Restaurant Operations Management at Algonquin College.
Website: https://paranoidpeoplelivelonger.com