teacher

teacher is a portfolio project that demonstrates a full Retrieval-Augmented Generation (RAG) application with:

• a FastAPI backend
• a PostgreSQL vector store
• an OpenAI-powered answer generation flow
• a Streamlit frontend for testing the /ask endpoint

---

Demo

---

Project Structure

frontend/ Streamlit UI
backend/ FastAPI API, ingestion pipeline, database schema, and RAG logic

---

Features

• Wikipedia-based dataset generation
• Retrieval-Augmented Generation (RAG)
• Vector search with PostgreSQL
• FastAPI-based API
• Streamlit demo frontend
• Docker-based local development

---

Quick Start

Backend

cd backend  
make build  
make up  

Swagger UI:
http://localhost:8000/docs

---

Frontend

cd frontend  
make build  
make up  

Streamlit UI:
http://localhost:8501

---

Demo Flow

1. Ingest documents into the vector database
2. Open the Streamlit frontend
3. Send a question to the /ask endpoint
4. Review the generated answer and sources

---

Documentation

• See backend/README.md for backend details
• See frontend/README.md for frontend details