RAG-Powered Document Assistant

Overview

Built a Retrieval-Augmented Generation system for question answering over academic PDF documents ### Implementation Details • Built a Retrieval-Augmented Generation system for question answering over academic PDF documents • Implemented semantic chunking, vector embeddings, and search with Sentence Transformers and ChromaDB • Integrated a local LLM via Ollama to generate grounded, citation-backed answers with verbatim page references

Key Features & Architecture