System Requirements Document (SRD)

Project Name: polar-ai

1. Introduction

The polar-ai project aims to develop an advanced AI-powered online examination proctoring system. This system is designed to address the growing need for secure and fair online examination environments, particularly in the context of remote education. By leveraging cutting-edge AI technologies, polar-ai will monitor and analyze user behavior during exams to detect and flag suspicious activities, ensuring integrity and fairness.

This document outlines the system requirements for polar-ai, incorporating advanced features such as computer vision-based face detection, eye tracking, behavioral analysis, real-time webcam and audio monitoring, tab-switching detection, automated cheating detection, secure authentication, and data security. These updates reflect the latest technological advancements and the specific needs of educational institutions in India and globally.

2. System Overview

polar-ai is an intelligent proctoring system that integrates AI technologies to monitor and detect cheating during online exams. The system will cater to diverse educational institutions, including schools, colleges, and certification platforms.

Key features include:

• Real-time monitoring of user behavior during exams using AI-powered tools.
• Detection of suspicious activities such as tab switching, unusual browser activity, and rapid question switching.
• Advanced computer vision capabilities for face detection and eye tracking to analyze user focus and behavior.
• Behavioral analysis to identify patterns indicative of cheating.
• Webcam surveillance and audio monitoring for comprehensive exam oversight.
• Secure authentication mechanisms to verify user identity.
• Robust data security measures to protect user privacy and comply with regulations.
• Compatibility with existing examination platforms or standalone deployment.

The system will be designed to ensure scalability, adaptability, and compliance with local regulations in India and globally.

3. Functional Requirements

Story Points:

• As a School Admin, I should be able to integrate the system into my institution's existing examination platform.
• As a College Admin, I should be able to monitor real-time exam activity and receive alerts for suspicious behavior.
• As a Certification Platform Admin, I should be able to generate detailed reports on flagged activities post-exam.
• As a User (Student), I should be able to take exams without intrusive monitoring while ensuring fairness.
• As a System, I should be able to detect multiple logins from the same account during an exam.
• As a System, I should be able to analyze rapid question switching patterns to flag potential cheating.
• As a System, I should be able to optionally use facial recognition to verify the identity of the exam taker.
• As a System, I should be able to track browser activity to detect unusual behavior during exams.
• As a System, I should be able to monitor webcam and audio feeds in real-time to detect suspicious activities.
• As a System, I should be able to analyze eye movements and facial expressions to identify potential cheating behavior.
• As a System, I should ensure secure authentication mechanisms to verify user identity before and during exams.
• As a System, I should provide robust data security to protect user information and comply with regulations.
• As a System, I should automate cheating detection using AI-powered algorithms for efficiency and accuracy.

4. User Personas

Admin:

• School Admin: Responsible for integrating the system into the school's examination platform and ensuring compliance with institutional policies.
• College Admin: Oversees exam monitoring and handles flagged activities during exams.
• Certification Platform Admin: Manages post-exam analysis and reporting for certification purposes.

User:

• Student: Takes exams through the platform while adhering to fair practices.

System:

• AI Proctoring System: Monitors exams, detects suspicious activities, and ensures compliance with security and privacy standards.

5. Visuals Colors and Theme

Color Palette:

• Background: #E8F1F5 (Gentle Ice Blue)
• Surface: #FFFFFF (Pure White)
• Text: #2C3E50 (Midnight Navy)
• Accent: #FF5733 (Fiery Orange)
• Muted Tones: #AAB7C4 (Soft Steel Gray)

This palette reflects a modern, professional aesthetic with a focus on clarity and trust, suitable for educational institutions and proctoring systems.

6. Signature Design Concept

The polar-ai homepage will feature a futuristic AI-powered exam monitoring interface that immerses users in the proctoring experience.

Key Features:

• Interactive AI Proctor Dashboard: A sleek dashboard with real-time visualizations of exam activity, including flagged behaviors, heatmaps, and live monitoring feeds.
• Behavioral Analysis Animation: A dynamic animation showcasing how the system analyzes facial expressions, eye movements, and audio patterns to detect cheating.
• AI Avatar Guide: A virtual AI assistant that interacts with users, explaining system features and offering tips for fair exam practices.
• Webcam Feed Integration: A live webcam feed preview with AI-driven annotations highlighting detected behaviors (e.g., "Eye movement detected").
• Tab-Switching Alerts: A visual indicator that flashes whenever a user switches tabs during an exam.
• Micro-Interactions: Smooth transitions, glowing icons for active monitoring tools, and hover effects to enhance user engagement.

This design concept ensures the homepage is visually striking, informative, and memorable, creating a lasting impression on users.

7. Non-Functional Requirements

• The system must ensure data privacy and comply with GDPR and Indian IT regulations.
• The system should support scalability to handle large numbers of concurrent users.
• The system must provide high availability with a minimum uptime of 99.9%.
• The system should offer multi-language support, including English and regional Indian languages.
• The system must ensure compatibility with major browsers and devices.
• The system should provide low-latency real-time monitoring and analysis.

8. Tech Stack

Frontend:

• React for Web

Backend:

• Python
• FastAPI

Database:

• MySQL or MariaDB (using Alembic for migrations)

AI Models:

• GPT 5.4 for user-friendly responses
• Claude 4.6 Opas for academic or coding work
• Gemini 3.1 Pro for friendly responses
• Google Nano Banana for image generation

AI Tools:

• Litellm for LLM Routing
• Langchain

Local Orchestration:

• Docker
• Docker-compose

Server-side Orchestration:

• Kubernetes

9. Assumptions and Constraints

Assumptions:

• The system will primarily be used by educational institutions in India, but it should support global deployment.
• Users will have access to stable internet connections during exams.
• Institutions will provide necessary permissions for advanced monitoring features like facial recognition and audio tracking.

Constraints:

• The system must comply with local and international data privacy laws.
• Facial recognition and movement tracking features must be optional to accommodate privacy concerns.
• Integration with existing platforms may require custom APIs.

10. Glossary

• AI: Artificial Intelligence
• GDPR: General Data Protection Regulation
• LLM: Large Language Model
• Docker: A platform for containerized application deployment
• Kubernetes: An orchestration system for managing containerized applications
• Tab Switching Detection: Monitoring user activity to identify when they switch tabs during an exam

This updated SRD incorporates all requested features and enhancements for **polar-ai**, ensuring it meets the needs of educational institutions while leveraging cutting-edge AI technologies. Let me know if further refinements are needed, Sujatha!