---
title: Customer Support RAG Assistant
emoji: 💬
colorFrom: blue
colorTo: green
sdk: gradio
app_file: app.py
---

# Customer Support RAG Assistant

A Gradio demo for a customer-support Retrieval-Augmented Generation (RAG) pipeline.

## Components
- Sentence Transformers for embeddings
- FAISS for FAQ retrieval
- LangGraph for the workflow
- Groq API for hosted language-model generation
- Gradio for the web interface

## Configure the Groq API key
In the Space, open **Settings → Secrets** and add:

- Name: `GROQ_API_KEY`
- Value: your Groq API key

Optionally add `GROQ_MODEL` as a Space variable if you want to select a different model enabled for your Groq account.

Do not put API keys in source code or upload them in `.env`.

## Notes
The first question may take longer because the embedding model and FAISS index are initialized on first use. The language model itself is accessed through Groq.