RAG System and NL2SQL Agent

Combining Retrieval-Augmented Generation with SQL capabilities to power enterprise analytics.

Overview

This project integrates a Retrieval-Augmented Generation (RAG) pipeline with a natural language-to-SQL (NL2SQL) agent. The goal is to allow non-technical users to query enterprise databases and document collections in plain English.

RAG Project Demo

Key Features

  • Context-aware document retrieval using vector search (FAISS).
  • SQL Agent powered by LangChain and OpenAI.
  • BigQuery integration with view support.
  • Access to structured and unstructured sources.

Tech Stack

Python, LangChain, OpenAI API, FAISS, Google BigQuery, Slack API (for interface), Terraform (infra setup).

Impact

This system streamlined decision-making by eliminating the dependency on technical staff for data queries. It also demonstrated the synergy between LLMs and classic databases in production environments.