GJU Student Regulations Assistant

The description states that GJU Student Regulations Assistant is a bilingual, citation-first tool built to help students find answers in official university regulations by making large PDF documents searchable. The author, Ahmad Dlahmea, built it using Python and Streamlit, with Supabase for data storage and OpenAI embeddings for retrieval. It indexes four official GJU regulation documents and supports English and Arabic queries, returning answers with citations to specific articles and PDF pages.

The system uses hybrid retrieval (vector, keyword, section-aware), article-aware chunking, and OCR fallback for Arabic text. It enforces read-only access via row-level security and includes automated tests and GitHub Actions for deployment.

Key commercial signals are absent: no revenue, customers, pricing, or traction data are provided. The tool is described as a prototype submitted to a hackathon, with no indication of ongoing use or adoption beyond the author's own testing.

The description states that the GJU Student Regulations Assistant is a bilingual, citation-first assistant designed to help students locate answers in official university regulations. It allows users to ask questions in English or Arabic and returns answers grounded in official documents.

It uses:

• Vector search with pgvector
• Full-text and section-aware retrieval
• Reranking of evidence
• GPT-5.6 for answer generation
• Supabase PostgreSQL as backend
• Streamlit for the frontend

The tool indexes four official PDFs from GJU’s Laws & Regulations page, including 2026 English and Arabic regulation books.

It is described as a public-facing application, accessible via a Streamlit app at https://gju-regulations-assistant.streamlit.app/, with read-only access enforced by Supabase row-level security.

Inference The product is a search-and-citation tool for university regulations, not a general-purpose AI assistant or marketplace. It is built as a prototype and deployed via GitHub Actions and Streamlit.

The description states that the tool was inspired by the challenge of students needing quick answers from large, scattered PDFs of university regulations. The author emphasizes that correct answers must include citations to the exact regulation, article, and PDF page, not just explanations.

It positions itself as:

• A searchable version of official documents
• Not a replacement for official sources
• Grounded in official university texts
• Designed to reduce friction in finding accurate information

The author claims that the tool makes regulations searchable without replacing the official sources, and that it provides reliable citations.

Inference The positioning is narrow: focused on improving access to official university rules, not on broader AI or education market opportunities. It is a utility for students and staff, not a commercial product.

The description states that the tool is built for GJU students, who often need quick answers from official university regulations.

It also mentions that it helps:

• Students find answers quickly
• Provide citations to exact articles and PDF pages
• Support both English and Arabic users

There is no indication of other target personas beyond GJU students. The tool is not described as targeting faculty, staff, or other institutions.

Inference The ICP (Ideal Customer Profile) is limited to GJU students, with a potential expansion to university staff or departments in the future.

The description does not provide any evidence of:

• Revenue streams
• Pricing model
• Monetization strategy
• Customer acquisition costs
• Subscription or usage fees

It is described as a public-facing prototype, deployed via Streamlit, and available at no cost to users.

Inference No business model is evident. The tool appears to be a non-commercial prototype, not a product with a monetization plan.

The description states that the system:

• Is built using Python and Streamlit
• Uses Supabase PostgreSQL with pgvector
• Employs OpenAI embeddings for vector search
• Uses GPT-5.6 for answer generation (and OCR fallback)
• Processes PDFs with PyMuPDF, including OCR when needed
• Preserves article and page metadata through ingestion, chunking, and retrieval
• Enforces read-only access using Supabase row-level security
• Runs automated tests and GitHub Actions workflows

It is described as:

• A lightweight deployment that does not require GPUs or local models
• A version-safe system, with checksum-based updates
• Capable of handling Arabic reading-order issues

Inference The technical stack is functional for a prototype, but lacks enterprise-grade scalability or robustness. It is built with open-source and cloud tools, suggesting low-cost, self-contained development.

The description states that the tool:

• Is a working public prototype
• Was submitted to the OpenAI 2026 hackathon
• Has been tested by the author and others
• Includes automated tests and GitHub Actions
• Supports English and Arabic
• Indexes four official GJU regulation documents

There is no evidence of:

• User adoption or usage metrics
• Customer feedback or engagement
• Revenue or monetization
• Product iteration beyond the prototype stage

Inference The product is at a prototype maturity level, with no traction data or user base reported.

The description does not provide any information about:

• Competitors in the student regulation assistant space
• Similar tools or platforms
• Market size or competitive dynamics

It is described as a unique solution for GJU students, but no comparison to existing tools is made.

Inference No competitive context is evident. The tool appears to be a novel local solution, not part of a broader market.

• No commercial traction or adoption: The tool is described as a prototype submitted to a hackathon, with no evidence of real-world usage.
• No monetization strategy: No pricing, revenue, or business model is evident.
• Limited scope: It only supports GJU students and four regulation documents; no expansion plan is described.
• Dependency on official sources: The tool relies entirely on GJU’s official documents, which may not be updated or maintained consistently.
• No third-party validation: No independent verification of the system's accuracy or utility.

Inference The project is a non-commercial prototype, with no evidence of product-market fit or scalability beyond its initial use case.

1. How many GJU students are currently using this tool, and how do you know?
2. Has the university officially endorsed or adopted this tool for student support?
3. What is the plan to expand beyond GJU and its current documents?
4. Are there any plans to monetize or scale this product?
5. How is the accuracy of citations verified, and what happens when regulations are outdated or disputed?
6. Has there been feedback from university staff or students on usability or reliability?