Project Info

DishDash

Inspiration

We wanted to make meal planning fun and effortless—like dating apps, but for food. With so many recipes out there, DishDash helps users discover meals they'll truly love.

What it does

DishDash is a swipe-based recipe discovery app. Users swipe right to save dishes, and our AI learns their preferences to recommend better recipes over time. Users can also add their own recipes and auto-generate tags using Claude.

How we built it

We built the app in SwiftUI. Recipes are loaded from a local recipes.json file and rendered in a Tinder-style interface. Users can upload their own recipes with an image or URL, and Claude AI generates smart tags to improve search and personalization. All saved data and images are stored in the app’s Documents directory.

Challenges we ran into

Managing file I/O and distinguishing between bundled vs. user-generated content Displaying consistently sized cards regardless of image content or description length Ensuring stable integration with the Claude API Working with SwiftUI layout quirks and image caching

Accomplishments we're proud of

A clean, swipeable UI that feels intuitive Fully working authoring experience with image import and tag generation Smart recommendation engine that adapts to user swipes Persistent local storage without requiring internet connection

What we learned

How to use SwiftUI’s layout system with custom components Safe file handling in the iOS sandbox REST API integration with a language model (Claude) Optimizing user flows for both browsing and content creation

What's next

Cloud storage and user login for syncing across devices Support for dietary filters, cuisine types, and ingredient-based search AI meal planning and grocery list generation Publishing on the App Store Created at UC Berkeley AI Hackathon

Analysis

Compare with all teams

View

Metric

UpsideDownTurtle
2

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only

SwiftIn code

1 of 1 appear in the indexed code.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

21 KB

Source files

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.