DiscovAI-search - AI Search Tool Featuring Advanced Vector Matching and Efficient Caching

DiscovAI-search: A Comprehensive Project Overview

DiscovAI-search is an advanced AI-powered search engine designed to simplify the discovery of AI tools and manage your data more effectively. Through innovative technology and a user-centric design, DiscovAI-search provides users with precise, reliable, and intuitive search capabilities.

💻 Live Demo

For those eager to experience DiscovAI-search firsthand, a live demo is available at DiscovAI.io. You can try it for free without needing to sign up or provide credit card details.

🗂️ Overview

Below is a concise guide to understanding DiscovAI-search, its features, tech stack, quickstart instructions, and deployment options.

🛠 Features

Vector-based Search: This feature transforms user queries into vectors, enabling highly accurate similarity matching across a comprehensive AI product database. It ensures users find exactly what they are looking for with minimal effort.
Redis-powered Caching: By leveraging Redis for caching, DiscovAI-search dramatically increases response speed. This ensures that recurring queries are managed efficiently, saving users time and providing a smoother experience.
Comprehensive AI Database: DiscovAI-search maintains a vast and regularly updated AI product database that spans numerous categories and industries, offering users a broad spectrum of tools to explore.
LLM-powered Responses: Utilizing large language models, DiscovAI-search provides detailed and context-aware responses that help users understand and evaluate search results.
User-friendly Interface: The interface is designed for ease of use, ensuring smooth navigation and making the discovery of AI products straightforward and enjoyable.

🧱 Tech Stack

DiscovAI-search is built using a robust and modern technology stack to ensure reliability and performance:

App framework: Built with Next.js, a React-based framework known for its flexibility and ease of use.
Text streaming: Utilizes Vercel AI SDK to provide seamless text processing.
LLM Model: Employs gpt-4o-mini to enhance AI interaction capabilities.
Database: Uses Supabase for managing comprehensive datasets with efficiency.
Vector: Integrates Pgvector for vector processing.
Embedding Model: Powered by Jina AI for precise and effective embedding tasks.
Redis Cache: Uses Upstash for reliable caching solutions.
Component library: Shadcn/ui provides extensive UI components.
Headless component primitives: Radix UI offers core modalities without restrictive preset styling.
Styling: Styled using Tailwind CSS, ensuring a modern and responsive design.

🚀 Quickstart

You can get started with DiscovAI-search effortlessly through the following steps:

Clone the Repository: Use Git to clone the repository with git clone https://github.com/DiscovAI/DiscovAI-search.
Install Dependencies: Navigate into the project directory and run pnpm i to install necessary dependencies.
Setup Supabase: Create a new Supabase project and run the provided SQL setup script to initialize your database.
Configure Upstash: Follow the Upstash guide to set up your Redis instance and acquire the necessary URL and token.
Fill Out Necessary Secrets: Copy the example environment configuration and set up your .env.local file with all required keys and tokens.
Run Locally: Start your local server with pnpm dev and access DiscovAI-search at http://localhost:3000.

🌐 Deploy

Deployment of DiscovAI-search can be done easily on platforms like Vercel, Zeabur, or Cloudflare Pages, allowing for flexible hosting solutions that meet varied business needs.