Searching Assistant

Inspiration

The goal of the Searching Assistant is to make searching for posts more accurate and tailored to users' specific needs. By integrating AI-based filtering, we aimed to create a tool that not only retrieves relevant content but also helps users connect with the right authors and find context-specific information faster and more efficiently.

What it does

The Searching Assistant allows users to search and filter social media posts more effectively by leveraging AI-based filtering to provide context-specific, relevant results. It helps users quickly find the most pertinent content and connect with the right authors based on their specific search needs.

How we built it

The Searching Assistant was built using the following architecture: Frontend: We used React to build a dynamic and user-friendly interface for the Chrome extension. React helps efficiently manage the UI and interactions, allowing for a seamless search and filtering experience. Framework: We leveraged WXT, a next-gen web extension framework, to streamline the development process. WXT provided advanced features to optimize extension performance, improve compatibility, and manage communication between the extension and web pages. Backend: To optimize the search results, we integrated the Gemini API, which uses AI to filter posts based on user intent and specific search criteria.

Challenges we ran into

Challenges We Ran Into Versioning Issues: During development, we faced problems related to version mismatches in dependencies, especially when working with Chrome’s Manifest V3 and integrating external libraries. Facebook Post Scraping: Another significant challenge was scraping posts from Facebook. The social media platform’s frequent updates to its privacy policies and post formats made it difficult to get consistent and clean data for analysis.

Accomplishments that we're proud of

Resolving Version Issues: We successfully addressed the versioning problems by updating dependencies and carefully managing compatibility between the Chrome extension APIs and libraries used. Content Scraping: We overcame the challenges of scraping content from social media by setting up custom scrapers and managing data flow efficiently. AI Integration: We managed to connect with the Gemini API, which significantly improved search optimization, making results more context-aware and relevant.

What we learned

Web Scraping: We gained hands-on experience with scraping data from social media posts. This required setting up custom scraping scripts to extract relevant content. Chrome Extension Development: We learned how to use Chrome’s extension APIs to interact with the browser and manage permissions. React Integration: We explored how to use React to create a dynamic and responsive UI for the extension. React allowed us to easily build reusable components and efficiently manage state across the application. WXT Framework: We became familiar with WXT, a next-gen web extension framework, which significantly simplified building the extension by providing modern tools and optimizations for handling web extension-specific challenges.

What's next for Searching Assistant

If possible, we would like to expand the tool to include scraping for Facebook, as the platform lacks an AI-powered search engine for finding specific job or housing opportunities. This would enable users to apply advanced filtering and search capabilities to more tailored content on Facebook.

Built With

• html
• javascript
• react
• tailwind
• typescript
• wxt