RegTrace

The description states that RegTrace is an AI-powered investigation aid for Canadian aviation professionals. It allows users to describe operational problems in plain language and returns a structured regulatory assessment based on a 44-document corpus including the Aeronautics Act, Canadian Aviation Regulations, and Transport Canada manuals.

Key features include:

• Fact extraction from user input
• Clarifying questions when facts are missing
• Investigation planning
• Search across multiple regulatory documents using purpose-built tools
• Distinguishing between binding and advisory material by authority tier
• Returning assessments (supported, conditional, insufficient, or conflicting)
• Deep-linking to exact sections of source documents
• Producing printable RegTrace Case Files

The system is built on a Next.js frontend, Supabase backend, and uses GPT-5.6 via OpenAI Responses API with function calling for evidence retrieval.

Inference The product appears to be a proof-of-concept or prototype built as part of a hackathon submission, not yet deployed in production at scale.

The author positions RegTrace as an extension of FlightRef, which was previously developed as a side project for searching aviation regulations. During Build Week, the author expanded it into a structured investigative tool rather than just a search engine.

Claims made:

• RegTrace turns operational problems into "structured, citation-gated regulatory investigations."
• It is not a legal determination but an aid to help teams build defensible trails back to official text.
• The system distinguishes between Acts, Regulations, Standards, Advisory material, and Guidance by authority tier.
• Users can correct facts or add details and see updated assessments without restarting the process.

Inference This evolution from a search tool to an investigative assistant suggests intent to move beyond simple information retrieval toward decision support in regulated environments.

The description states that RegTrace focuses on Canadian airlines and operators, specifically groups such as flight crews, dispatchers, and flight standards teams who regularly reference regulations during operations.

These users are described as needing to:

• Reference regulations while dealing with operational scenarios
• Search by keywords rather than regulation numbers
• Identify missing facts, determine binding sources, reconcile provisions, and preserve defensible trails

Inference The target customer is likely internal staff within regulated aviation organizations — not external clients or consumers. The ICP seems to be mid-to-senior-level professionals in regulatory compliance or operational roles.

There is no evidence provided about pricing, monetization, or business model. The description does not mention any revenue streams, subscriptions, licensing fees, or customer acquisition costs.

Not evidenced

The system uses:

• GPT-5.6 via OpenAI Responses API
• Function calling with four bounded tools: regulation search, exact-section reading, related-document lookup, and definition lookup
• PostgreSQL full-text search combined with pgvector search in Supabase
• Next.js 14, TypeScript, React, Tailwind CSS, Vercel
• Zod for input/output validation
• SSE streaming for real-time interface updates

The author reports using Codex extensively during development to implement tasks like orchestration, testing, and QA.

Inference The technical stack indicates a modern web application built with AI integration, likely intended for demonstration or prototyping rather than enterprise deployment. Use of function calling and tool grounding suggests an attempt at controlling output quality.

The description states:

• Live end-to-end investigations completed in 34–53 seconds during QA
• All 35 unique citation links from rehearsals opened the intended source sections
• A clear repository history separating FlightRef foundation from RegTrace work
• The product was submitted to an OpenAI hackathon

However, there is no evidence of:

• Actual users or customer base
• Revenue or monetization
• Product adoption or usage metrics
• Production deployment or scaling efforts

Not evidenced

The description does not provide any information about competitors or market positioning beyond the author’s own claims. No mention is made of existing tools for regulatory compliance, aviation regulation search engines, or AI-powered legal research platforms.

Not evidenced

• Unverified claims: All functionality and performance are self-reported without independent validation.
• Prototype nature: The product appears to be a hackathon submission with no evidence of real-world use or scalability.
• AI dependency: Heavy reliance on GPT-5.6 raises concerns about consistency, cost, and control over outputs.
• Limited scope: Focus on Canadian aviation regulations may limit broader applicability unless expanded.
• No commercial traction: No evidence of revenue, customers, or market validation.

Inference Without external verification or data on adoption, the risk of misalignment between stated capabilities and real-world utility is high.

1. What specific regulatory domains outside of Canadian aviation are you planning to support?
2. How do you plan to validate that model outputs align with official interpretations of regulations?
3. Have you tested RegTrace with actual users from the aviation industry?
4. Are there any known limitations or edge cases in how the AI handles ambiguous or conflicting regulations?
5. What is your roadmap for moving beyond a prototype into a scalable product?
6. How do you intend to monetize this tool, if at all?
7. Can you demonstrate how the citation-gating mechanism prevents false references from appearing in results?