PantryPal

Inventory Page
Recipe Page
Inventory Page 2
Recipe Page 2
Recipe Page 3
Home Page

Inspiration

Food insecurity affects many students at UC Davis, and The Pantry ASUCD provides essential free ingredients to combat this. The issue is young college students, many being away from home for the first time, don't know how to turn these raw ingredients into actual meals.

We wanted to empower students with a simple, intuitive platform which helps them get the most out of their pantry hauls. Our goal was to build something directly impactful for our campus and community.

What it does

The Pantry at ASUCD — Recipe Finder is a web application that helps students discover meals they can make using ingredients from the Pantry.

Users can

Browse pantry inventory by category, dietary tags, or availability
Select ingredients they picked up from the Pantry
Instantly receive recipe matches ranked by how well they use selected ingredients
See exactly what ingredients they have vs. what they still need
Get AI-powered personalized recipe suggestions
View generated images for recipes to make them easier to visualize

How we built it

We built this as a full-stack web app using modern tools with scalability in mind. Our tech stack consisted of

Next.js and React for building a fast, responsive interface.
Supabase to store recipes, ingredients, and pantry inventory
A relational schema connecting recipes and ingredients for efficient matching
OpenAI GPT-4o-mini selects the best matching recipes from our real database and formats them cleanly
DALL·E generates images for suggested recipes to enhance usability and engagement

Challenges we ran into

One of our biggest challenges was incorporating AI in a way that produced consistent, practical results rather than unpredictable outputs. We had to coordinate effectively as a team under tight time constraints. Connecting the backend to the frontend required careful handling of API routes, data formatting, and state management to keep the application responsive. Additionally, we had to think beyond the prototype stage and consider scalability, structuring the code in a way that could support future growth and additional features.

Accomplishments that we're proud of

We’re proud that we built a fully functioning, end-to-end application that addresses the real campus issue of food scarcity. From ingredient input to AI-generated recipe output, the entire system works cohesively, and it demonstrates how technology can directly enhance community resources rather than replace them.

What we learned

Through building The Pantry at ASUCD — Recipe Finder, we learned how to integrate a frontend, backend, and AI system into one cohesive, production-ready application. Incorporating AI taught us the importance of clear prompt design and structured data flow for reliable outputs. We also strengthened our ability to collaborate under time pressure, divide responsibilities effectively, and maintain clean version control. Most importantly, we gained experience quickly writing organized, scalable, industry-standard code.