QuerySense

An AI chatbot using Retrieval-Augmented Generation (RAG) to answer user questions based on custom internal documents. Relevant document chunks are retrieved and passed to an LLM, which generates contextual responses.

Traditional search struggled to return relevant, context-aware answers from large internal datasets. We needed a scalable way to understand both documents and user intent.

We used embedding models to vectorize document chunks and user queries, stored them in a pgvector-enabled PostgreSQL database, and integrated a large language model to formulate final responses.