StudyBuddy Local: Your Private AI Study Partner. Your Notes Stay Yours.

Description and use

StudyBuddy Local is an offline, privacy‑first study assistant that ingests PDFs/TXT, creates local embeddings, retrieves relevant passages, and generates grounded answers, quizzes, and summaries using a local LLM, all without cloud dependencies.

Problem

Students need a private, offline study assistant that can transform lecture notes into chat, quizzes, and summaries without uploading sensitive documents to the cloud.

Outcome

The tool enables fully offline, privacy-preserving study assistance, providing grounded answers with citations, quizzes, and summaries without cloud dependencies.

Workflow

• Upload PDF/TXT to local Streamlit interface
• PyMuPDF extracts text and page metadata
• Chunker splits content into sentence-aware overlapping chunks
• Sentence Transformers generates embeddings on-device
• ChromaDB stores embeddings locally
• Student asks a question via chat interface
• Retrieval matches question to relevant chunks and returns passages
• Ollama with Qwen2.5 3B generates answer with source citations
• App displays answer, quiz, or summary
• Pydantic validates quiz output
• Pytest tests document processing, retrieval, chat, and quiz flows

Evidence

StudyBuddy Local is a private study partner I built for a friend preparing for university exams.

Source

Extracts text and page metadata from PDFs and text files, then splits it into overlapping, sentence-aware chunks.

Source

Creates embeddings on-device and stores them in a local ChromaDB collection.

Source

Generates validated multiple-choice quizzes, summaries, and revision notes using a local Ollama model (Qwen2.5 3B by default).

Source

Keeps the interface bound to localhost, disables Streamlit usage statistics, and has no cloud fallback.

Source