Inspiration
Meal planning is one of the most repetitive and time-consuming tasks families face every week. At the same time, food waste continues to be a major global problem. Many households throw away perfectly good ingredients simply because they don’t know what to cook with what they already have.
I wanted to build a system that combines AI, automation, and personalization to reduce food waste while simplifying weekly food decisions for families.
What it does
AI Menu Planner is an AI-powered system that automates weekly meal planning and adapts to real family needs.
The app can:
- Generate a fully personalized weekly meal plan
- Regenerate individual meals instantly
- Regenerate the entire week (Pro feature)
- Automatically build and synchronize a smart shopping list
- Generate recipes based only on ingredients available at home (Pantry Mode)
- Support multiple family members with different dietary preferences and calorie targets
The system keeps everything connected — when meals change, the shopping list updates automatically.
How we built it
The project is built using a scalable AI-driven architecture:
- Mobile app built with Expo (React Native + TypeScript)
- Backend powered by .NET 8 Minimal API
- Azure OpenAI for structured recipe and meal generation
- Asynchronous background processing for weekly regeneration
- Supabase (Postgres + Auth) for data storage
- Azure Container Apps for deployment and scalability
- RevenueCat for subscription management
Weekly regeneration runs asynchronously in the backend, allowing the UI to remain responsive while AI jobs are processed in the background.
Challenges we ran into
Some of the main challenges included:
- Designing reliable structured outputs from AI (ingredients, steps, nutrition)
- Handling asynchronous generation without breaking user experience
- Keeping shopping lists synchronized with regenerated meals
- Managing personalization logic across multiple family members
- Balancing AI quality, performance, and cost
Accomplishments that we're proud of
We’re proud of building a system that feels cohesive and production-ready rather than just a simple AI demo.
Key accomplishments:
- Fully working async AI generation pipeline
- Real-time synchronization between meals and shopping list
- Flexible regeneration at both meal and weekly level
- A clean and intuitive mobile UX
- A clear monetization model integrated into the experience
What we learned
This project reinforced that building AI products is not just about calling an API — it requires architectural thinking, user workflow design, and system synchronization.
We learned how to structure AI outputs, manage long-running background jobs, and design a responsive mobile experience around asynchronous AI generation.
What's next for AI Menu Planner
Next steps include:
- Improving AI personalization per family member
- Adding smarter nutrition balancing
- Expanding Pantry Mode with ingredient recognition
- Enhancing performance and generation speed
- Launching publicly on the App Store
AI Menu Planner is designed as a foundation for intelligent food automation — helping families save time, reduce waste, and eat better with AI.
Built With
- .net
- expo.io
- nativewind
- openai
- react
- rest
- revenuecat
- supabase
- typescript
Log in or sign up for Devpost to join the conversation.