SavingFood.ai
Frontend for food waste reduction workflows
What it is
A frontend experience for helping food service teams understand spoilage risk, track waste, and make better inventory decisions. The interface turns model outputs and operational data into clear dashboards for busy operators.
The Problem
Food waste data is only useful when people can understand it quickly and act on it. The product needed to make spoilage signals, waste trends, and inventory guidance feel practical rather than like another complex analytics tool.
My Role
Focused on the frontend implementation: dashboard layouts, waste tracking views, inventory-oriented summaries, prediction result states, responsive interaction patterns, and the visual language for communicating operational insights.
Approach
Built the frontend with React and Next.js, shaping a clean dashboard experience around prediction results, trend visualization, and actionable inventory information. The UI prioritizes clear status, readable data summaries, and quick paths for users working on smaller screens.
Hardest Decision
Making the predictions useful without requiring extensive data input — finding the balance between prediction accuracy and the practical data entry burden on users who are often busy food service operators.
Status
Live at sf-project-topaz.vercel.app as a frontend experience for spoilage prediction, waste tracking, and inventory optimization workflows.
What I'd Change
Would add integration with point-of-sale systems for automatic waste tracking, implement computer vision for food quality assessment, and add community sharing features for surplus food redistribution.