Inspiration

We wanted a simple tool that helps students actually use their lecture PDFs—ask questions, review key ideas, and study more efficiently without digging through slides manually.

What it does

LectureNav lets students upload a PDF, ask questions about the content, generate summaries, create practice questions, and view a clickable history of all interactions.

How we built it

We built a Spring Boot backend that extracts PDF text, creates local embeddings, retrieves relevant chunks, and sends them to a Cerebras LLM. The frontend is one static HTML/JS page that communicates with the backend via fetch().

Challenges we ran into

Handling messy PDF text, chunking the lectures cleanly, tuning embedding similarity, managing LLM context, and keeping the frontend simple but responsive.

Accomplishments that we're proud of

A fully working end-to-end system, fast local embedding search, clean UI, and reliable AI responses for Q&A, summaries, and practice questions.

What we learned

How to integrate embeddings with an LLM, handle PDF processing, design clean prompt pipelines, and keep a minimal frontend connected to a structured backend.

What's next for LectureNav

Multi-PDF support, improved UI, flashcard/quiz modes, cloud deployment, and optional offline LLM support for faster, local-only usage.

Built With

Share this project:

Updates