Building Astuto Next

Building Astuto Next is an innovative project focused on improving product feedback management. The platform aims to simplify the sharing, discovery, and management of user suggestions by consolidating them into a single space. This eliminates the issue of duplicated ideas and ensures that users can easily find existing suggestions that match their proposals.

Astuto Next provides several key features, including company feedback boards, feature requests, community voting, an AI-powered duplicate detector, and a unique Halloween-inspired interface. Users are able to submit ideas by providing a title and description, enabling others to understand the problem and proposed solution more clearly. Moreover, the platform allows users to vote on existing ideas, ensuring that popular requests receive attention and consideration from companies.

A notable aspect of this project is its use of AI to detect duplicate feature requests. Users can leverage AI to check if a similar suggestion already exists before submitting a new one. This is accomplished using the Ollama model (Llama 3.2 1B) for local AI integration. The platform's backend is built using Node.js and Express.js, while the frontend is constructed with HTML, CSS, and JavaScript. Data management is handled by SQLite using the better-sqlite3 library.

Throughout the development process, the author encountered various challenges, such as integrating the frontend with backend API routes, creating and managing database tables, and handling API errors. However, these obstacles fostered valuable learning experiences in frontend-backend communication, database management, and AI integration. The author emphasizes that the AI duplicate detection feature required extensive testing and fine-tuning rather than a simple AI-generated output.

Looking ahead, Astuto Next has the potential for further enhancements, with plans to include AI-powered feedback analysis to identify common topics, notifications for status changes in feature requests, improved company interaction mechanisms, and enhanced duplicate detection accuracy. The author expresses enthusiasm for receiving feedback on potential AI-powered features that could further enrich the public feedback platform.