Paris Bauer

Built a work-search and application system to match opportunities against what the candidate can actually do. The system hard-screens roles, researches employers that survive screening, maps evidence and genuine gaps, and generates applications tailored to the specific employer and role while keeping factual employment history unchanged. Human review is required for every application, with the ability to approve, edit, or reject before sending. Approved applications become live CVs, and the system records observable post-send behaviour through CV Intelligence, treating evidence conservatively and avoiding claims the evidence cannot establish.

Building PAi OS, a human–AI operating system being developed with PAi Training as its first proving environment. I define product and system thinking: what problem is solved, what the system should and should not do, how human judgement and AI work together, how evidence flows into decisions, and how the product is tested and simplified as it evolves. I established the relationship between the PAi OS engine and the PAi Training product for real human-context testing, and I specified system behaviour including decision authority, evidence flows, edge cases, failure handling, and longitudinal learning needs. I work with AI and code to turn product decisions into functioning software, test behaviour, identify failures, and iterate based on evidence while maintaining clear ownership and simplicity.

Designed and iterated a live application system combining evidence-based role matching, readable information presentation, and engagement tracking. Reviewed AI-assisted outputs for accuracy, relevance, clarity, and faithful representation of real experience; revised the system when evidence revealed errors.