StudySphere-AI

Inspiration

Students often struggle to find specific information in lengthy textbooks, PDFs, and lecture notes. Traditional learning methods require students to search through multiple pages manually, making studying time-consuming and less efficient.

We were inspired to build StudySphere-AI, an AI-powered learning companion that helps students interact with their study materials, discover relevant information, and move toward a more personalized learning experience.

What it does

StudySphere-AI is an AI-powered educational platform that helps students search and learn from their uploaded study materials.

Key features include:

Document Upload: Upload supported educational documents, including PDFs and PowerPoint presentations.
Intelligent Document Processing: Extract text and organize content into manageable chunks.
Semantic Search: Find relevant passages using sentence-transformer embeddings and FAISS vector similarity search.
Context-Aware Learning: Retrieve relevant information from uploaded documents to support answers grounded in learning materials.
Relevant Context Retrieval: Identify when the available content does not sufficiently support a query.
Web-Based Learning Interface: Access learning resources through a user-friendly web application.

Our goal is to make studying more efficient, accessible, and interactive.

How we built it

We built StudySphere-AI using a combination of web technologies, document-processing tools, and AI-powered retrieval techniques.

Python and Flask: Backend application and API development.
HTML, CSS, and JavaScript: User interface and frontend interactions.
SQLite: Storage for learning materials, document metadata, and extracted chunks.
PyPDF: PDF text extraction.
Sentence Transformers: Generation of semantic embeddings for document content and queries.
FAISS: Efficient similarity search across embedded document chunks.
Retrieval-Augmented Generation (RAG): A workflow designed to combine retrieved document context with AI-generated explanations.

The application processes documents, divides their content into chunks, generates embeddings, and indexes them for semantic retrieval. When a student asks a question, the system searches for relevant passages that can support a context-aware response.

Challenges we ran into

Developing StudySphere-AI involved several technical challenges:

Document Processing: Extracting and organizing text from uploaded educational documents.
Chunking Strategy: Dividing documents into meaningful sections while preserving useful context.
Semantic Retrieval: Improving the relevance of search results

Built With

ai
css
faiss
flask
git
html
javascript
natural-language-processing
pypdf
python
rag
sentence-transformers
sqlite

Submitted to

Multimodal AI Hackathon 2026

Created by

Frontend and User Experience
UI/UX development
Responsible for the Flask/Jinja2 interface, responsive layouts, materials page, tutor interface, progress dashboard, accessibility improvements, and user interaction.

Lahari Kummari
Testing, Integration and Documentation
Quality and delivery
Responsible for integrating modules, testing application routes and workflows, validating citations and error handling, maintaining project documentation, and preparing the working-model demonstration.

Sakshitha Ayyala
AI Tutor and RAG Pipeline
AI and retrieval
Responsible for semantic search, embeddings, FAISS-based retrieval, Retrieval-Augmented Generation (RAG), AI Tutor responses, and source-grounded citations.

Lasyasaimanasa Jujjavarapu
Document Processing and Ingestion
Backend development
Responsible for PDF and PPTX processing, text extraction, document chunking, timestamped TXT transcript ingestion, and preparing learning materials for indexing.

Chevula Vidhya