Norma - Compliance and Procedure Compiler

The description states that Norma is a policy and procedure compiler. It allows users to upload PDF documents (e.g., policies, regulations) and uses GPT-5.6 to interpret them into structured proposals. These proposals are kept separate from executable policy until human review and approval.

After approval:

• A questionnaire is generated for evaluating cases.
• The evaluation is performed by a deterministic rule engine, which ensures consistent outcomes based on inputs.
• Each result includes:
  • Overall outcome
  • Passed/failed requirements
  • Missing information
  • Manual review conditions
  • Source references supporting findings

The system supports exporting results in JSON and Markdown formats.

Inference This implies that Norma functions as both an AI-assisted interpretation tool and a compliance workflow platform — not a chatbot or general-purpose document assistant.

The description states that Norma was built with the idea of turning policies and procedures into traceable, human-reviewed assessments, aiming to reduce manual effort in compiling regulations into operational workflows.

It positions itself as:

• A compliance compiler — not a chatbot.
• A system where AI proposes, but humans approve.
• Designed around the principle: “AI proposes. Humans approve. Deterministic software evaluates.”

The authors also claim that Norma is built for regulated environments, where auditability and traceability are critical.

Inference The positioning suggests Norma targets industries with strict regulatory requirements (e.g., finance, healthcare, public administration), where human oversight and audit trails are essential.

The description does not explicitly name target customers or define an Ideal Customer Profile (ICP). However, it implies that the product is intended for:

• Organizations managing complex policies or procedures.
• Users who need to convert policy documents into actionable workflows.
• Teams requiring human review and auditability in compliance processes.

It also mentions a fictional student-support scholarship policy as a demo use case, suggesting potential relevance to educational institutions or government agencies.

Inference Based on the context of regulated environments and human review workflows, Norma likely targets enterprise clients in sectors like public administration, legal services, or compliance-focused industries.

No information is provided about pricing, monetization strategy, or business model. The description only describes how the product works technically and conceptually.

Inference Since no revenue or customer data is shared, it's unclear whether Norma intends to be sold as a SaaS product, a tool for internal use, or part of a larger platform offering.

The system is described as a full-stack TypeScript application built with Next.js, React, and Tailwind CSS. It uses:

• GPT-5.6 via OpenAI Responses API
• Structured Outputs
• PDF.js for handling documents
• Docker for deployment
• Prisma for database access
• Auth.js for authentication
• Vitest for testing

Key technical features include:

• Document ingestion and text extraction
• Source verification (citations validated against original document)
• Human review workflow
• Deterministic rule engine with three-state logic (true, false, unknown)
• Export capabilities in JSON and Markdown
• Private workspaces and data isolation
• Reproducible deployment via Docker

Inference The architecture shows a clear separation of concerns between AI interpretation, human review, and deterministic execution — which may indicate scalability and auditability.

There is no evidence of traction or maturity beyond the hackathon demo. The project was built in one week and deployed publicly as a demonstration. No mention of:

• Revenue
• Customers
• User base
• Product adoption
• Market feedback
• Iteration history

Inference Norma is currently at a prototype or proof-of-concept stage. There is no indication of real-world usage or product-market fit.

No explicit competitors are mentioned in the description. However, based on the functionality described (policy compilation, AI interpretation, human review, deterministic evaluation), Norma could compete with:

• Compliance automation tools
• Workflow management platforms for regulated industries
• Document analysis and rule engine systems

It also overlaps conceptually with:

• AI-powered document processors (e.g., Notion, DocuSign, PandaDoc)
• Regulatory compliance software (e.g., LogicGate, ComplyAdvantage)

But no direct comparison or competitive positioning is made.

Inference Norma’s unique value proposition lies in combining AI interpretation with deterministic execution and traceability — a niche but potentially valuable approach for regulated environments.

1. No traction or commercial validation: The product exists only as a hackathon demo.
2. Unproven market demand: No evidence of customer interest, use cases, or feedback.
3. Dependency on GPT-5.6: Reliance on a single AI model raises risk if access changes or pricing increases.
4. Limited scope in demo: The demo uses a fictional policy; real-world complexity may not be fully addressed.
5. No business model disclosed: Unclear how the product will generate revenue or scale.

Inference Without real-world testing or monetization plans, Norma remains unproven in terms of viability and scalability.

1. What specific industries or use cases are you targeting beyond the demo?
2. How do you plan to validate demand for this product in regulated environments?
3. What is your strategy for scaling beyond a single hackathon project?
4. Are there any existing partnerships, pilot programs, or early adopters?
5. How will you handle model dependency (e.g., GPT-5.6) long-term?
6. What are the key assumptions behind the deterministic evaluation approach?
7. Do you have plans to integrate with other compliance tools or platforms?