Type
AI-powered Online Exam Proctoring Platform
Objective
One of the leading HRtech companies in the United States sought to develop an automated online exam proctoring platform. Their main goal was to ensure secure, reliable, and scalable remote assessments while maintaining academic integrity. The solution needed to support a high volume of concurrent users and offer seamless proctoring with minimal manual intervention.
Technologies
Java, Python, AWS, Docker, Kubernetes, ReactJS, SQL
Country
Belgium
Project Attributes
Engagement Model
Dedicated Team Engagement
Duration
14 months
App Users
Educational institutions, exam administrators, and students
Challenges
- Ensuring Secure Proctoring: The client needed a platform that could prevent cheating and detect any
suspicious behavior during online exams, all while maintaining user privacy. This
required advanced AI-based surveillance features integrated with the platform. - Scalability for High Traffic: The system needed to support thousands of students taking exams
simultaneously across multiple regions. Ensuring the platform could scale
effectively while providing seamless performance was crucial. - Manual Monitoring Overload: Exam administrators faced challenges in manually monitoring every test-taker.
The client sought an automated system that could assist in proctoring while
reducing the dependency on manual intervention. - User Experience and Accessibility: The platform needed to be user-friendly, ensuring smooth navigation for students
and administrators. Any issues with platform usability could lead to exam
disruptions and dissatisfaction.
Solutions
- AI-Based Proctoring System:
Adorebits developed a robust AI-driven proctoring system capable of monitoring
test-takers in real-time. The system used computer vision and facial recognition
technologies to track suspicious activities such as gaze deviation, unauthorized
persons in the room, and attempts to access external resources. - Cloud Infrastructure for Scalability:
Leveraging AWS, Docker, and Kubernetes, Adorebits created a cloud-native
solution that could automatically scale to accommodate thousands of concurrent
users. This ensured high availability and minimal latency, even during peak exam
times. - Automated Cheating Detection:
The proctoring system was equipped with an AI-based algorithm that flagged
potential cheating incidents. The platform would automatically alert
administrators to review suspicious cases, drastically reducing the manual
workload involved in monitoring each exam. - User-Friendly Interface and Accessibility:
Built with ReactJS, the platform provided an intuitive and responsive interface for
both administrators and students. Comprehensive testing was conducted to
ensure accessibility across various devices and operating systems, providing a
seamless experience for users.
Results:
- Enhanced Security and Integrity: The platform’s AI-based proctoring system ensured that exams were conducted securely, detecting suspicious behavior in real-time. Academic integrity was maintained without invading user privacy.
- Scalability Achieved: The platform successfully scaled to handle thousands of concurrent test-takers without any performance issues, thanks to the cloud infrastructure built on AWS, Docker, and Kubernetes.
- Reduced Manual Proctoring: The AI-driven system reduced the need for manual intervention by 50%, allowing administrators to focus on reviewing only flagged cases instead of monitoring every test-taker manually.
- Improved User Experience: Both students and administrators reported a 30% improvement in platform usability, with fewer technical issues during exams and easier navigation of features.
Conclusion:
Adorebits’ AI-driven online exam proctoring platform delivered on the client’s objectives of secure, scalable, and user-friendly remote exams. By automating crucial aspects of proctoring and leveraging cutting-edge cloud technologies, Adorebits provided a future-ready solution for educational institutions conducting online assessments.