AI Semantic Search API 🧠

AI-powered semantic search API built with FastAPI, OpenAI Embeddings, and ChromaDB.

This project demonstrates the core concepts behind:

• Embeddings
• Semantic Search
• Similarity Search
• Vector Databases
• RAG (Retrieval-Augmented Generation)

---

Features 🚀

• Store documents with embeddings
• Semantic similarity search
• Vector database integration
• Metadata filtering
• FastAPI REST API
• ChromaDB persistent storage

---

Tech Stack 🏗️

Technology	Purpose
FastAPI	Backend API
OpenAI Embeddings	Semantic vector generation
ChromaDB	Vector database
Python	Core language

---

Project Structure 📁

app/
├── main.py
├── config.py
├── schemas.py
│
├── services/
│   ├── embedding_service.py
│   └── vector_store.py

---

Installation ⚙️

Clone Repository

git clone https://github.com/your-username/ai-semantic-search-api.git

Move Into Project

cd ai-semantic-search-api

Create Virtual Environment

Windows

python -m venv venv
venv\Scripts\activate

Linux / Mac

python3 -m venv venv
source venv/bin/activate

---

Install Dependencies 📦

pip install -r requirements.txt

---

Environment Variables 🔑

Create .env file:

OPENAI_API_KEY=your_openai_api_key

---

Run Server 🚀

uvicorn app.main:app --reload

Open Swagger Docs:

http://127.0.0.1:8000/docs

---

API Endpoints 📚

Method	Endpoint	Description
GET	/	Health check
POST	/documents	Add document
POST	/search	Semantic search
GET	/documents	List all documents
DELETE	/documents	Delete all documents

---

Example Document 📄

{
  "title": "FastAPI Tutorial",
  "content": "FastAPI is a modern Python framework for building APIs quickly.",
  "category": "backend"
}

---

Example Search 🔍

{
  "query": "How can I build API with Python?",
  "limit": 3
}

---

Semantic Search Example 🧠

Query:

How can I build APIs with Python?

Result:

FastAPI Tutorial

Even without exact keyword matching.

This works because embeddings capture semantic meaning.

---

Concepts Practiced 📖

This project helps practice:

• Embeddings
• Vector Search
• Similarity Search
• Cosine Similarity
• Semantic Retrieval
• Vector Databases
• RAG Foundations

---

Future Improvements 🚀

• PDF Upload Support
• Text Chunking
• PostgreSQL + pgvector
• User Authentication
• AI Chatbot Integration
• RAG Pipeline
• Redis Caching
• Hybrid Search

---

Learning Goals 🎯

This project is designed for developers learning:

• AI Engineering
• RAG Systems
• LLM Applications
• Vector Databases
• Semantic Search

---

License 📄

MIT License