Study Pilot: An AI-Powered Platform for Personalized Learning through Retrieval-Augmented Generation on Diverse User Content
1st Ravi Teja Karnati
Dept. of Computer Science Engineering,
Guru Nanak Institutions Technical Campus Hyderabad, India
ravitejakarnati5312@gmail.com
2nd Prof. (Dr.) Harish Kundra
Dean of Faculty and Student Affairs,
Guru Nanak Institutions Technical Campus, Hyderabad, India
drhkundra@gniindia.org
3rd P R S Santosh Naidu
Dept. of Electrical and Communication Engineering,
Guru Nanak Institutions Technical Campus Hyderabad, India
prssn.raja@gmail.com
Abstract—Personalized learning is a critical goal in modern education, yet existing digital platforms often struggle to adapt to the unique needs and materials of individual learners. Many AI-driven educational tools rely on static datasets, limiting their relevance to a user’s specific curriculum or personal study resources. This paper presents Study Pilot, a comprehensive AI- powered platform designed to revolutionize personalized learning by enabling dynamic interaction with diverse content uploaded directly by the user, including PDFs, web links, and YouTube videos. At its core, Study Pilot employs Retrieval-Augmented Generation (RAG) [7], integrating advanced LLMs (via Ollama [1] and Groq API) with a vector database (ChromaDB) and other techniques to transform heterogeneous sources into co- hesive, interactive learning experiences. The platform supports multi-source course creation, contextual chat grounded in user content and the web, multiple language support via translation, collaborative course sharing, discovery of user-created resources, community forums, enhanced podcast generation for human- like audio [2], and a course rating system. Built with a Python Flask backend [3] and a JavaScript-based frontend [5], Study Pilot demonstrates a robust and scalable approach to making advanced, adaptable learning tools accessible to a wide range of users and content types. We present the system’s architecture and methodology, and initial evaluations of the core document- based RAG functionality show promising results in accurate and relevant information retrieval for personalized learning.
Index Terms—Personalized Learning, Educational Technology, Retrieval-Augmented Generation, Large Language Models, AI in Education, Adaptive Learning, Multi-modal Learning, Collabo- rative Learning.