OpiqBot & OpiqGPT

About Opiq

Opiq is a cloud-based learning platform widely used in public schools across several countries, offering over 500 digital textbooks and teacher resources that align with national curricula. The platform caters to students, teachers, and parents, providing a comprehensive range of tools for learning, teaching, and monitoring progress. With its interactive and engaging content, Opiq enhances both classroom and home-based learning experiences. The platform also supports a broad range of subjects, making it an essential tool for modern education.

The Challenge

Opiq, a widely used digital education platform across several countries, sought to enhance personalized learning by integrating advanced predictive analytics into its system. The goal was to predict and improve student performance, particularly focusing on 7th grade mathematics. This involved building a sophisticated, low-latency machine learning (ML) pipeline capable of real-time predictions. The challenge was multi-dimensional: it included leveraging data from the "Opiq Dashboard," validating a hypothesis through initial linear regression models, and eventually developing a production-grade ML pipeline hosted on AWS Sagemaker. Additionally, the project required real-time event aggregations using Kafka and Redis, ensuring the system could deliver predictions within seconds of student interactions. The system also needed to integrate seamlessly with Opiq’s existing platform, maintain high standards of data privacy, and incorporate a recommendation engine to dynamically adjust the difficulty of exercises based on the student’s performance.

The Result

Bitropia successfully implemented an advanced machine learning solution that transformed Opiq into a more dynamic and responsive educational platform. The system enables real-time predictions of student performance, initially focusing on 7th grade mathematics, with future expansions planned. By utilizing AWS Sagemaker for model training and hosting, Kafka for real-time data aggregation, and Redis for low-latency storage, the platform delivers personalized feedback almost instantly.

Furthermore, the collaboration with Tallinn University resulted in developing a recommendation engine that tailors exercises to the student’s current learning session, offering a balanced mix of challenges and easier tasks. This personalized approach helps optimize the learning process and improves student outcomes. Extensive testing in schools across multiple regions validated the system's efficacy, making Opiq a leading example of data-driven, personalized education. The success of this project not only enhances the current platform but also sets the stage for future developments in other subjects and educational markets.