All work
AINutritionClaude API

Macro Chef

AI-Powered Meal Planning & Nutrition Tracker

Nutrition is where most body recomposition programs fail. People know they need to hit their macros but don't know what to cook. Macro Chef is an AI-powered meal planner that takes your exact macro targets, dietary restrictions, and available ingredients, then generates recipes that actually fit. Three modes (Quick Meal, Meal Prep, Pantry Raid) for different needs on Tuesday night vs. Sunday afternoon. I built it because the #1 question from our gym community was 'What should I eat to hit my goals?' and nothing on the market answered it.

Visit Live Site
Macro Chef

The Problem

Knowing what to eat to match your specific fat loss and muscle gain goals is left entirely to guesswork. MyFitnessPal tracks food but doesn't generate plans. Generic meal plans fail because everyone has different dietary constraints, preferences, and goals. The result: people know their macro targets but have no idea what to cook to hit them, and the friction of manual logging means most give up within a week.

Research & Discovery

1

Talked to members of Everfit Motion (our gym community), and most tracked nutrition in spreadsheets or not at all. The gap wasn't motivation, it was that existing tools (MyFitnessPal, Cronometer) don't generate plans matched to your specific goals

2

The #1 question from gym members was 'What should I eat to hit my goals?' Meal planning matched to specific macro targets and body recomposition phases was the killer feature they couldn't find anywhere else

Strategy

Target customer: People pursuing body recomposition who know their macro targets but struggle with what to actually eat. Competitive landscape: MyFitnessPal tracks food but doesn't generate plans. Cronometer is detailed but doesn't connect to goals. No product generates personalized meal plans matched to specific body composition targets. Product thesis: The AI meal planner is the differentiator: nutrition matched to your specific goals is the hardest problem to solve and the one most people give up on. Current status: Testing with Everfit Motion gym community.

What I Built

AI Meal Planning Engine

Nutrition is where most recomposition programs fail. People know they need to hit their macros but don't know what to cook. The AI planner takes your exact macro targets, dietary restrictions, and available ingredients, then generates recipes that actually fit. Three modes (Quick Meal, Meal Prep, Pantry Raid) because users have different needs on Tuesday night vs. Sunday afternoon.

AI Meal Planning Engine
AI Meal Planning Engine
AI Meal Planning Engine

Meal Prep & Shopping

Meal prep is the bridge between planning and execution. The planner lets you set meals per day, servings per recipe, and max prep time, then generates a full week of meals with a consolidated shopping list. Items are grouped by recipe so you know exactly why each ingredient is on the list.

Meal Prep & Shopping
Meal Prep & Shopping
Meal Prep & Shopping

Personalization & Preferences

Generic meal plans fail because everyone has different constraints. The preference system captures dietary restrictions, cuisine preferences, spice tolerance, preferred proteins, food dislikes, cooking skill, and health goals, all fed into the AI prompt so every recipe is personalized. Macro targets are fully editable so the app grows with your goals.

Personalization & Preferences
Personalization & Preferences
Personalization & Preferences

Key Decisions

1

Chose Claude over GPT for meal planning because of longer context windows. The prompt includes full macro targets, dietary restrictions, available ingredients, and recent meal history

2

Replaced barcode scanning with plain-text AI extraction after dogfooding showed the friction of manual logging killed adherence

3

Three distinct planning modes instead of one generic planner because the use case is fundamentally different when you need dinner in 20 minutes vs. prepping meals for the week

Results & Impact

  • AI meal planning generating nutritionally-targeted recipes in under 3 seconds via Claude API
  • Meal logging time reduced from 5 minutes to 30 seconds with plain-text AI extraction
  • Full personalization system ensuring every generated recipe matches user constraints and preferences

Tech Stack

Next.jsTypeScriptSupabaseAnthropic Claude APITailwindCSS