MenuTaste AI Agent

MenuTaste is a web-based enterprise AI agent for food entrepreneurs, cafes, restaurants, and small food businesses. It helps users turn a food or drink idea into a structured product-quality, nutrition, risk, market-fit, and launch-readiness report. The user enters a product name, ingredients, business type, location, customer segment, dietary focus, target price, and preparation complexity. MenuTaste then runs a multi-step workflow: it performs local ingredient and nutrition-signal analysis, detects allergens and dietary conflicts, scores the product across nutrition, quality, market fit, and operations, and sends a structured prompt to a Featherless open-source model for deeper business reasoning. The final output includes an executive summary, nutrition estimate, risk review, ingredient notes, Featherless AI reasoning, recommended actions, launch checklist, and downloadable Markdown and JSON reports. This makes MenuTaste useful for early-stage restaurants and food entrepreneurs who need to validate a menu item before piloting it with customers. The system combines deterministic local scoring with Featherless AI reasoning. This hybrid design keeps core food-risk checks explainable while using open-source LLM inference for strategy, recommendations, customer positioning, and launch planning. The app is designed as a production-style Streamlit web application and deployed on Vultr for the hackathon demo.