MacroMeter

Overview

A nutrition tracking web application with natural language meal entry, powered by Firebase.

Key Features

• Natural language meal parsing with smart quantity recognition using regex pattern matching for complex meal inputs
• Real-time Firebase Cloud Functions backend serving as serverless proxy to USDA FoodData Central API with 500k+ foods
• Offline-first architecture with localStorage persistence for daily meals, goals, favorites, and 30-day rolling history
• Customizable daily goals with real-time progress dashboard featuring color-coded visual feedback
• Complete meal management system with favorites, history tracking, and CSV export for data portability

Challenges & Solutions

The project tackled complex challenges including natural language parsing ambiguity with sequential USDA database searches and partial success handling, API rate limiting mitigation through sequential processing with 300ms delays and exponential backoff retry logic, nutrition data normalization across different USDA database versions using fallback field matching, cross-domain API access via Firebase Cloud Functions CORS proxy, localStorage data synchronization across multiple buckets with date-based separation for daily resets, and portion quantity calculations with proper handling of fractional amounts and natural language quantities.