🌱 QueryMind: AI-Powered Knowledge Assistant

QueryMind is an intelligent, context-aware Q&A assistant built with Streamlit, LangChain, Google Gemini, and FAISS.
It enables users to create a custom knowledge base from CSV data and retrieve precise, context-driven answers using Retrieval-Augmented Generation (RAG).

---

🚀 Overview

QueryMind transforms structured CSV data into a searchable knowledge base and uses Google Gemini to generate contextually accurate answers.
It demonstrates seamless integration of:

• LangChain RAG pipelines
• Gemini 2.5 Flash model
• FAISS vector search
• Hugging Face embeddings
• Streamlit UI for interactivity

---

🧠 Core Features

• 🧾 CSV Knowledgebase Creation — Converts your CSV into embeddings using Hugging Face’s Instructor-large model.
• 🔍 Semantic Retrieval — Uses FAISS to find the most relevant context.
• 🤖 Gemini-Powered QA — Generates factual, grounded answers using Gemini 2.5 Flash.
• 🧩 LangChain Integration — Implements document and retrieval chains for efficient RAG workflows.
• ⚡ Streamlit UI — Interactive, minimal, and intuitive interface.
• 🔒 Secure Setup — API key management using .env.

---

🧩 Tech Stack

Category	Tools / Frameworks
Frontend	Streamlit
Backend (LLM Integration)	LangChain
Language Model	Google Gemini 2.5 Flash
Vector Store	FAISS
Embeddings	Hugging Face Instructor-large
Env Management	Python-dotenv
Language	Python 3.10+

---

⚙️ How to Run the Project

Follow these simple steps to get QueryMind running locally 👇

1️⃣ Clone the Repository

git clone https://github.com/Shanu06github/QUERYMIND.git

2️⃣ Navigate to the Project Directory
cd QueryMind

3️⃣ Add Your Google API Key
GOOGLE_API_KEY=your_google_api_key_here

4️⃣ Run the Application
streamlit run main.py

5️⃣
Once the app starts, open the displayed local URL (usually http://localhost:8501) in your browser.

Try out few sample questions from the csv file (filewithresponse.csv)