HackStack

Project Info

Cuisine Alchemy

Devpost

Inspiration

The culinary world is one steeped in tradition. There's beauty in several-generations-old recipes, classic local ingredients, and learning how to cook from your grandma. With rapid globalization, the culinary world is also one that has innovated at every turn - big bold flavors, melting pots of cultures, and food fairs. People from opposite ends of the world connect and bond through our fundamental love for food! Our goal is to join these aspects of food and use it to connect foodies!

What it does

Users can pick between chef personas from different countries. Two LLM agents are then separately prompted, and each of them represent one of the two chosen chefs. The agents build off each other, brainstorm different ideas, and reason about how varied ingredients' taste profiles would interact. Finally, end up with an innovative idea, DALL-E generated image, and a recipe for a brand new dish combining the two cuisines and cultures. We also allow the user to generate their own persona for the LLM agent to assume. The user can enter their cultural and cuisine background, their favorite recipes, and any dietary restrictions. The user-generated agent can then proceed to interact with any of the other agents to come up with new dish ideas tailor-made to the user's own culture! To share their new-found food combinations, we generate a tweet, recipe, and image of the new dish for users to share on their social media.

How we built it

We built the LLM agents using OpenAI API and LangChain. We built the front-end using next.js, Tailwind CSS and the back-end with Flask. We used DALL-E 3 to generate images. What we accomplished One of the most challenging parts of this project was innovating in this unique sector of food, AI and technology. We are most proud of being able to simulate convergent dialogue between two artificially intelligent agents.

Analysis

Compare with all teams

View

Metric

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
  • CSSIn code
  • FlaskIn code
  • JavaScriptIn code
  • LangChainIn code
  • Next.jsIn code
  • OpenAIIn code
  • PythonIn code
  • ReactIn code
  • Tailwind CSSIn code
  • TypeScriptIn code

10 of 10 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

31 KB

Source files

23

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

0 stars