Fridgeo
Fridgeo is a SvelteKit MVP for an AI recipe finder that combines community recipes, fridge scanning, swipeable AI ideas, friends, chat, and recipe sharing.
Stack
- SvelteKit + TypeScript
- Firebase client setup for auth, Firestore, and storage
- Gemini server route for recipe generation
Setup
- Install dependencies:
npm install
Update environment values in
.env.Start the app:
npm run dev
Notes
GEMINI_API_KEYis kept server-side and used only insrc/routes/api/ai/recipes/+server.ts.- Firebase web config stays client-side through
VITE_FIREBASE_*variables. - The UI is wired for Firebase data sources; until you connect them, lists will render empty states.
Inspiration
Reducing food waste and the daily "what should we cook" decision. Fridgeo turns what you already have into quick, shareable meal ideas.
What it does
- Scans a fridge photo or ingredient list and generates AI recipes.
- Lets users swipe through recipes, save favorites, and share with friends.
- Provides community recipes, chat, and profile-based collections.
How we built it
- SvelteKit + TypeScript for the UI and routing.
- Firebase for auth, Firestore, storage, and real-time features.
- Gemini API for ingredient understanding and recipe generation.
Challenges we ran into
- Balancing rich UI interactions (swipes, animations, scroll video) with performance.
- Keeping auth state consistent across routes and deployments.
- Designing a clean, mobile-friendly layout with many features.
Accomplishments that we're proud of
- A cohesive end-to-end flow from fridge scan to shared recipe.
- Swipe-first AI experience with real image support.
- A unified profile hub for saved, liked, and friend features.
What we learned
- Small UX details make or break trust in AI outputs.
- Clear auth errors and feedback improve onboarding.
- Progressive enhancement keeps the app usable even with missing data.
What's next for Fridgeo
- Full Firebase data integration for community content and chat.
- Better recipe image sourcing and nutrition data.
- Smarter personalization (diet filters, cuisine preferences, and history).
Built With
- firebase
- geminiapi
- svelte
- typescript
Log in or sign up for Devpost to join the conversation.