Semantic Search & RAG for Intelligent Access to Company Knowledge

Python LangChain LangGraph Vertex AI Cloud Run BigQuery Docker Streamlit

I implemented an AI-powered knowledge system that utilizes Retrieval-Augmented Generation (RAG) and semantic search to make legacy, siloed, and previously undigitized information more accessible and discoverable through natural, semantic queries.


The system enables efficient retrieval across long-form documents such as contracts, construction plans, and other unstructured data that are not easily searchable using traditional keyword-based approaches.


The solution achieved a 50% improvement in search accuracy compared to the legacy keyword-based system previously in use.

RAG System Architecture RAG System Example

← Back to Projects