ReanMate. An AI Turtor and Private classroom.

Faculty of Technology

BSc (Hons) in Computer Science

ReanMate

Ai Tutor And Private Classroom

Abstract

        ReanMate is an AI-powered learning platform for people in Cambodia. The intended Minimum Viable Product (MVP) features teachers setting up private classes, inviting students to join online via a join code, uploading PDFs and videos to the class, enabling the creation of assignments and quizzes, and managing each class's student usage of AI assistance. Furthermore, students are able to enroll in authorized courses, view lessons created by the teacher, complete assignments and quizzes, and use AI tools to ask questions or get answers from lessons, quizzes, and assignments, all based on the teachers' uploaded content. This system is engineered to provide Khmer, English, or mixed-language questions, and to indicate the source(s) and the pages in which each AI answer is generated.

       This project illustrates advanced object-oriented programming concepts, such as: classes, service classes, repository classes, validation, composition and separation of responsibility. It shows the structure of a procedural system which allows to decouple routes/controllers, business services, repositories, models, templates, and database access together. Users, classes, memberships, lessons, videos, documents, assignments, conversations, messages, sources, practice questions and AI usage records are stored in a MySQL DB. The AI workflow involves processing, chunking, embedding retrieval of the PDFs by page along with a grounded prompt that asks the OpenAI API or Gemini for an answer.

Introduction

Background

        Digital learning systems is increasingly which combine course materials, assignments, communications, and personalized learning support. ReanMate proposes a focused learning environment in which teacher’s own lesson material become the main knowledge source for AI tutor instead of threating AI as a general chatbot, this platform connects the AI experience directly to private class and teacher also provided PDF lessons.

Problems Statement:

·      Students need additional explanations after class.

·      PDF materials alone do not provide interactive learning support.

·      General AI may provide answers outside the teacher's materials.

·      Students need answers they can verify from the original lesson.

·      Teachers need control over AI usage in their classes.

·      Private class materials must be protected from unauthorized access.

·      Khmer/English bilingual AI support is needed.

·      Teachers need classes, lessons, PDFs, and assignments in one platform.

·      AI usage must be controlled and monitored.

Solutions

           ReanMate will provides private classes, teacher-managed PDF lessons, assignments, and a controlled AI tutor that answers questions based on authorized class materials.

Objectives:

·     Implement secure login and role-based access.

·     Manage lessons and PDF learning materials.

·     Manage assignments and student progress.

·     Build a PDF-grounded AI Tutor.

·     Support Khmer and English learning.

·     Provide AI answers with source-page references.

·     Generate practice questions.

·     Track and control AI usage.

·     Apply secure OOP-based system architecture.

Research and Development Questions:

1.     How can object-oriented design be used to separate learning-domain responsibilities from database and web concerns?

2.     How can private class membership be enforced throughout normal requests and AI retrieval?

3.   How can teacher-uploaded PDF material be transformed into searchable chunks while preserving page information?

4.  How can a RAG workflow provide useful AI explanations while reducing unsupported answers?

5.      How can the system support Khmer, English, and mixed-language questions?

6.      How can AI usage be measured and limited during a small demonstration?

Scope

·      Student, teacher, and administrator roles.

·      Private classes with join codes.

·      Class membership management.

·      Teacher-created lessons and private PDF materials.

·      PDF viewing and page navigation.

·      Teacher-created and published assignments.

·      Student assignment dashboard and status filtering.

·      Class-scoped AI Tutor and Study Workspace.

·      AI quick actions: Explain in Khmer, Summarize, Generate Practice Questions, and Quiz Me.

·      Source document/page citation for grounded AI answers.

·      Practice sets and questions.

·      AI usage tracking and limits.

·      Administrative monitoring of users, classes, documents, and AI usage.

Limited

·      Live video classes.

·      Native mobile application.

·      Payments and subscriptions.

·      School-wide administration.

·      Public course marketplace.

·      AI voice tutor.

·      Advanced proctoring.

·      Full LMS grading engine.

·      Automatic high-stakes grading.

·      Additional AI providers before the OpenAI-based MVP is reliable.

Target Users and Stakeholder

Literature and Technology Review

     The project uses object-oriented programming to represent important concepts such as User, Class, Lesson, Document, Assignment, Conversation, and AI Usage. The goal is not to create classes only for appearance. Each class should have a clear responsibility and meaningful methods. Services coordinate business workflows, while repositories isolate database operations.

        Composition is preferred where one component depends on another, such as an AI service depending on a repository or usage tracker. Inheritance should only be used when the relationship is meaningful, such as role-specific behavior where it improves the design. Type hints, validation, and documentation should be used for public methods.

Python

      Python is the main programming language for the course implementation. Its class system, type hints, modules, exception handling, testing ecosystem, and web frameworks make it suitable for demonstrating object-oriented design.

Flask

        Flask is proposed as the web framework required by the Advanced OOP course. Flask routes/controllers receive requests, validate access, call service methods, and return HTML views or JSON responses. The route layer should not contain direct SQL queries or complex business rules.

PostgreSQL or Mysql

       PostgreSQL is used as the primary relational database for the course implementation. Relational tables represent users, classes, memberships, lessons, documents, assignments, conversations, messages, and AI usage. Primary keys, foreign keys, unique constraints, and appropriate data types are used to maintain data integrity.

Retrieval-Augmented Generation

      ReanMate uses a retrieval-augmented generation concept. Teacher PDFs are processed page by page, cleaned, divided into chunks, and represented for semantic retrieval. When a student asks a question, the application checks authentication, class membership, AI availability, and usage limits before retrieving relevant material. The retrieved context is then used to construct a grounded prompt for the AI provider.

   A key requirement is that retrieval must remain inside the student's authorized class/document scope. The AI system should not retrieve chunks from another private class. When the material does not sufficiently answer a question, the tutor should state that the material is insufficient rather than inventing a source or page.

 OpenAI API

      The product specification selects the OpenAI API as the MVP AI provider. The API key must remain on the backend and must never be exposed to the browser. AI requests, token usage, and estimated cost should be recorded for monitoring.

Proposed System Features

      This is a comprehensive list of proposed features for a Digital Learning Platform, structured by user role and core functionality.

Session Module:

·      Register

·      Login

·      Logout

·      Password management

Authorization Module:

·      Student role

·      Teacher role

·      Admin role

·      Role-based access control

User Management Module:

·      User list

·      CRUD

·      Search

·      User report

Class Management Module

·      Class list

·      CRUD

·      Search

·      Class report

Lesson & PDF Module

·      Lesson list

·      CRUD

·      PDF upload/delete

·      Search

·      Processing status

Assignment Module

·      Assignment list

·      CRUD

·      Search/filter

·      Assignment report

AI Tutor Module

·      Ask AI

·      PDF-based retrieval

·      Source/page citation

·      AI usage tracking

Practice Module

·      Practice list

·      CRUD

·      Search

·      Practice results

Admin Module

·      User management

·      Class monitoring

·      AI usage monitoring

·      System controls

Object-Oriented Programming Class Design

Service layer:

Repository Layer:

           Repositories provide database access without UI logic. Example repositories include UserRepository, ClassRepository, ClassMemberRepository, LessonRepository, DocumentRepository, AssignmentRepository, ConversationRepository, PracticeRepository, and AIUsageRepository. Parameterized SQL or a safe database abstraction must be used instead of concatenating user input into SQL.

Project Structure

project_name/
├── app/
│   ├── __init__.py              # Flask application factory
│   ├── config.py                # configuration classes
│   ├── extensions.py            # db, login_manager, bcrypt, etc.
│   ├── models/                  # domain/data models
│   ├── repositories/            # MySQL data access
│   ├── services/                # business rules / workflows
│   ├── routes/                  # Flask blueprints
│   ├── forms/                   # WTForms or validation helpers
│   ├── templates/               # Jinja2 pages
│   └── static/                  # CSS, JS, images
├── tests/
├── migrations/ or sql/
├── docs/
├── run.py
├── requirements.txt
└── README.md

SYSTEM Architecture

        The proposed request flow is Browser → Flask Routes/Controllers → Services → Repositories → PostgreSQL. The AI flow adds document processing and retrieval components: PDF Upload → Validation → Private Storage → Page Extraction → Chunking → Embedding/Indexing → Authorized Retrieval → Grounded Prompt → OpenAI API → Saved Answer + Sources.

Database Design

         The API owns every data access decision. PostgreSQL is durable state; Redis and MinlO solve separate operation problems.

Why this design:

·      DATABASE_URL and storage credentials stay in the backend environment only.

·      File bytes are not databases column and relational rows keep private object metadata

·      Postgre remains authoritative even when Redis or MinlO is temporarily unviable.

ERD Diagram 

Main Relation

·      One teacher can own many classes.

·      A class can contain many students through ClassMember.

·      A class can contain many lessons.

·      A lesson can have one or more associated documents depending on final implementation.

·      A document can contain many page-aware document chunks.

·      A class can contain many assignments.

·      An assignment can have progress records for many students.

·      A conversation belongs to a user and can be scoped to a class/lesson/document.

·      A conversation contains many messages.

·      An AI message can have multiple source references.

·      AI Usage records connect usage to a user and optionally a class.

Implementation Plan

User Interface  Student Experience  Teacher Experience.  Admin Experience

AI Processing Workflow

AI Response Rule

·      Teacher material is the primary evidence.

·      If the material is insufficient, the system clearly says so.

·      The AI must never fabricate a source page.

·      Explanations should be readable and appropriate for students.

·      Khmer, English, and mixed-language questions are supported.

·      Useful English technical terms may be preserved in Khmer explanations.

·      For assessed assignments, the tutor should prefer explanations, hints, examples, and step-by-step guidance rather than automatically providing final answers.

API / Route Plan

Security Design

·      Hash passwords securely.

·      Keep the OpenAI API key on the backend only.

·      Apply server-side role and ownership guards.

·      Validate resource access on every protected request to prevent IDOR.

·      Use private/signed document URLs where applicable.

·      Validate PDF MIME type and maximum file size.

·      Rate-limit authentication, upload, and AI endpoints.

·      Log administrative actions and AI errors without logging secrets.

·      Keep classes private by default.

AI Usage Controls

Testing

Security Testing

·      Student attempts to access another class by changing a URL ID.

·      Student attempts to access another class document directly.

·      Teacher attempts to edit another teacher's class.

·      User attempts to change their role through a frontend request.

·      AI endpoint is called above its rate limit.

·      Vector retrieval attempts to cross a class boundary.

·      Non-PDF and oversized uploads are submitted.

AI Quality Testing

·   Question is directly answered in the PDF.

·   Question is partially answered in the PDF.

· Question is absent from the PDF and the AI should state that the material is insufficient.

·    Khmer-only question.

·    English-only question.

·    Mixed Khmer/English question.

·  Student makes an incorrect assumption and the tutor corrects it using the source material.

·    Source page opens the correct PDF location.

Development Methodology and Timeline

Development Phase

Proposed Weekly Timeline

Risk, Ethics, and Privacy

Risk ad Mitigating

Privacy:

·      Classes are private by default.

·      Documents are not public URLs.

·      Student conversations are not exposed to other students.

·      Teacher analytics should prefer aggregate signals rather than reading private chats.

·      Administrative access to user content should be minimal and audited.

·      Technical logs should avoid secrets and unnecessary private content.

Ethical AI Use

       ReanMate is intended to support learning rather than replace the student's work. For marked assignments, the tutor should prefer explanations, hints, examples, and step-by-step guidance rather than automatically providing a final answer. The system should also communicate uncertainty when the teacher's material does not contain enough information to answer.

Conclusion

            ReanMate proposes a focused AI-assisted learning platform that combines private classes, teacher-controlled PDF materials, assignments, and a document-grounded AI tutor. The design directly supports the Advanced OOP course requirements by separating domain objects, services, repositories, routes, and database access. It also provides a practical R&D problem: how to make AI assistance useful while maintaining private class boundaries, source traceability, and controlled usage.

            The MVP is deliberately limited to approximately ten users so that the team can prioritize a reliable end-to-end proof. The core demonstration is complete when a teacher creates a private class, uploads a real lesson, creates an assignment, a student joins the class, studies the PDF, asks a useful question in Khmer or English, receives a grounded answer with the correct source page, and completes the learning flow without manual database intervention.

 Future Work

·      Flashcards and improved AI lesson mode.

·      Richer assignment submissions and a stronger progress page.

·      Announcements and due-date notifications.

·      DOCX/PPTX support.

·      Mock exams and weak-topic detection.

·      Teacher review workflow.

·      School/organization accounts and institution analytics.

·      AI voice questions, OCR, lecture transcription, personalized tutoring, study plans, and AI-generated diagrams.

The product plan recommends delaying payment systems, school administration, additional AI providers, and other broad LMS features until the core student and teacher value is proven.

Reference

[1] S. Sokcheat, “Advanced Object-Oriented Programming,” lecture notes, University of Technology and Entrepreneurship, Phnom Penh, Cambodia, 2026.

[2] GeeksforGeeks, “Object-Oriented Programming (OOPs) Concept in Java,” GeeksforGeeks, 2026. [Online]. Available: https://www.geeksforgeeks.org/object-oriented-programming-oops-concept-in-java/