AI-Powered Search

Overview

Toolhouse Search helps users find tools efficiently using AI-based search with FAISS and sentence-transformer embeddings.

Tech Stack

• Backend: FastAPI, FAISS, Sentence Transformers
• Frontend: React (TypeScript), Axios
• Database: FAISS (Vector Search)

Features

• AI-powered semantic search using FAISS & Sentence Transformers
• Handles partial word matches (e.g., "machine" finds "machine learning")
• Fast & efficient search using vector embeddings
• If are no results "No results found" message appears
• Press "Enter" to search
• AI-powered ranking ensures the most relevant tools appear first
• Easily extendable architecture (add more tools, new search methods, etc.)
• Modular backend design with FastAPI & FAISS

Installation & Setup

Clone the Repository

git clone https://github.com/yourusername/toolhouse-search.git

cd toolhouse-search

macOS

Backend Setup (FastAPI)

cd backend

python -m venv venv

source venv/bin/activate

pip install -r requirements.txt

uvicorn app.main:app --reload

Frontend Setup (React) - in another terminal tab

cd frontend

npm install

npm start

Example API Request

curl -X POST "http://127.0.0.1:8000/search" -H "Content-Type: application/json" -d '{"text": "machine learning"}'

Demo Video

gravacao-do-ecra-2025-02-12-as-004312_F2BZcjji.mp4