Pantry Pal

Inspiration

People waste the food they have at home because they forget what they already have in their fridge. I wanted to solve a simple question:

“What can I cook right now with what I already own?”

What it does

PantryPal is an AI kitchen assistant that turns a fridge, pantry, or receipt images into meals.

It helps users:

• Upload picture of fridge, pantry, or receipts
• Detect ingredients automatically using AI
• Generate recipes from available food
• Create shopping lists for the missing items
• Suggest what to cook before food expires

Optional features:

• Craving Mode (cuisine-based recipes)
• Global Kitchen Challenge (daily world cuisine theme)
• Pantry Social Circles (light social cooking groups)

How I built it

PantryPal is built as a full-stack AI system with a structured pipeline inspired by AWS architecture.

Flow: Image Upload → AI Ingredient Detection → Pantry Dashboard → Recipe Generation → Shopping List

I used Kiro’s spec-driven workflow to design this system before implementation, ensuring that all the components followed a very clear architecture.

This system separates AI tasks into modular steps:

• Ingredient detection
• Recipe generation
• Insight generation

Challenges I ran into

• Ensuring that AI outputs were useful without overestimating ingredients
• Keeping the system simple while still adding some advanced features
• Designing a smooth end-to-end flow for fast demos

Accomplishments that I'm proud of

• Built a working AI pipeline from image → meal suggestions
• Created a clean, usable pantry dashboard
• Designed gamified and social extensions without breaking the core usability
• Maintained a complete end-to-end flow that makes it feel like a real product

What I learned

• AI works best when broken into structured steps, not single big prompts
• Simplicity leads to better user experience than having too many features
• Clear product storytelling is just as important as the technical implementation

What's next for Pantry Pal

• Barcode and receipt scanning improvements
• Deeper personalisation based on the user's habits or preferences
• Stronger social cooking circles, social events like potlucks
• Potential integration with grocery delivery platforms, such as FairPrice, Cold Storage, etc

Built With

• amazonapigateway
• bedrock
• css3
• dynamodb
• eventbridge
• express.js
• html5
• interfont
• javascript
• lambda
• multer
• node.js
• react18
• s3
• vite