matakltm-code/car-shopping-assistant

An AI-powered car shopping assistant for the Ethiopian market, using Laravel, Vue.js, Laravel Nova, and Gemini API for efficient, embedding-based searches. Inspired by Joseph Kibur’s LinkedIn idea. # Car Shopping Assistant An AI-powered car shopping assistant tailored for the Ethiopian market, using Laravel, Vue.js, Laravel Nova, and the Gemini API to provide efficient, embedding-based searches for buyers and sellers. ## Overview This project was inspired by a LinkedIn post from Joseph Kibur, who proposed an AI-powered Car Shopping Assistant to streamline the car buying process by handling queries on fuel type, seating capacity, budget, and availability. Taking this idea as a challenge, I developed a prototype in just 2 hours with the help of ChatGPT, creating a full-stack web application that enables users to search for cars in the Ethiopian market efficiently using advanced AI embeddings. The platform allows car owners to list their vehicles, buyers to search based on various criteria (e.g., location, fuel type), and administrators to manage users and inventory via a dashboard. It leverages the Gemini API for semantic search, making it scalable and relevant for users. ## Why I Built This Project I built this project to address the tedious and time-consuming process of car shopping, particularly in the Ethiopian market, where buyers often struggle to find vehicles that match their needs (e.g., suitable for local road conditions, affordable in Ethiopian Birr, and available in specific cities like Addis Ababa or Dire Dawa). Joseph Kibur’s idea highlighted the potential for AI to simplify this process, and I saw an opportunity to create a practical solution while deepening my skills in full-stack development and AI integration. ## Tech Stack - **Backend**: Laravel (for API development, business logic, and routing), Laravel Nova (for admin dashboard) - **Frontend**: Vue.js (for interactive user interface), Tailwind CSS (for styling and responsive design) - **AI Integration**: Gemini API (for generating embeddings to enable semantic search) - **Database**: MySQL (via Laravel Eloquent ORM) - **HTTP Client**: Guzzle (for API requests to Gemini, with SS …