ChatPDF

I built a retrieval-augmented generation pipeline to answer questions from large PDF documents. The system used LangChain and the Gemini API together with vector retrieval to handle large-context document queries more effectively.

What I implemented

• Built a RAG pipeline using LangChain and the Gemini API.
• Queried online PDFs and retrieved relevant context for answering user questions.
• Used vector retrieval to bypass LLM context limits for large document understanding.

Why it was useful

This project gave me practical experience in modern LLM application design, information retrieval, and real-world document QA workflows.