Coding Buddy: An Adaptive AI-Powered Platform for Personalized Learning

Harsha Nutalapati, Sachin Velmurugan, Nestor Michael C. Tiglao · 2024

In today's fast-paced tech world, students often struggle to keep up with complex programming concepts. Traditional online learning platforms fail to adapt to individual learning speeds, leaving many students frustrated and falling behind. Our research introduces “Coding Buddy,” an innovative online learning system designed to solve this problem. Coding Buddy uses advanced technology to track how each student learns and adapts the course content accordingly. By analyzing factors like the time spent on each lesson, the types of mistakes made, and how often students ask for help, our system creates a personalized learning experience. Our study of 1,000 computer science students showed that 70% found traditional online courses too rigid. With Coding Buddy, 85% of students reported better understanding of complex topics. The system's ability to adjust difficulty levels in real-time led to a 40% increase in course completion rates. The key innovation is our use of AI-powered chat assistants and granular difficulty levels. This allows Coding Buddy to offer explanations in simpler language when needed, or provide more challenging content for advanced learners. Our results suggest that this personalized approach significantly improves learning outcomes in tech education. By addressing the individual needs of each student, Coding Buddy represents a major step forward in online programming education, potentially revolutionizing how we teach and learn complex technical skills.

Read the paper · More papers on PaperTik