Show HN: Food Decoder (github.com)

Food Decoder is an innovative open-source web application aimed at enhancing users' awareness of their food consumption. It employs AI models along with a comprehensive database from Open Food Facts to assist individuals in maintaining their dietary habits, bodybuilding plans, and overall health. Notable features include in-browser barcode scanning, which allows users to quickly pull up ingredient information without needing to download an app. For products without barcodes, users can upload a photo of the food label, utilizing Vision LLMs (Nvidia Nemotron Nano 12b v2) to automatically extract textual information. This application stands out in the AI/ML community by integrating advanced machine learning techniques to generate health scores and highlight nutritional attributes while flagging potential concerns like hidden additives and sugars. Its resilient backend, powered by FastAPI and hosted on Hugging Face, ensures seamless operation and high uptime. The user-friendly interface features modern design elements such as glassmorphism and 3D effects, making it both visually appealing and functional. Although Food Decoder primarily serves educational purposes, it offers a significant opportunity for developers and health-conscious individuals to engage with AI-driven nutrition analysis.