Inspiration
The idea for Shelf Smart was born out of a need for smarter, healthier, and more accessible home cooking solutions. We realized that many people face the daily challenge of trying to use the ingredients they already have in their kitchens while maintaining a healthy diet or sticking to specific dietary preferences. Inspired by the idea of reducing food waste and improving meal quality, we decided to create a tool that makes it easy to modify recipes and make the most of what's available in the pantry.
What it does
Shelf Smart is an AI-powered smart recipe modifier that helps users take any recipe and tweak it based on dietary preferences or health goals. Users can input a recipe, and Shelf Smart will suggest ingredient substitutions for healthier, vegan, gluten-free, or other dietary-friendly alternatives. It also provides a nutritional comparison between the original recipe and the modified version, helping users make informed decisions about their meals.
How we built it
We used a combination of Streamlit for building a user-friendly interface and MongoDB Atlas to store recipe data and user preferences. The AI model was built using Databricks Open Source, which allowed us to implement natural language processing (NLP) to understand recipes and suggest intelligent substitutions. We also integrated external nutritional databases to analyze the caloric and nutritional content of ingredients. For deployment and hosting, we utilized Cloudflare to ensure the app runs smoothly and securely.
Challenges we ran into
One of the main challenges was training the AI model to accurately understand and modify complex recipes. Handling multiple dietary preferences without sacrificing the integrity of the original recipe was tricky. We also had to ensure the nutritional calculations remained accurate, even when suggesting substitutions. Balancing simplicity with powerful functionality for a seamless user experience was another challenge.
Accomplishments that we're proud of
We’re proud of creating an intuitive, AI-driven tool that provides real value to users looking to improve their cooking habits. By building a functional prototype within a limited time, we were able to achieve our goal of making Shelf Smart both user-friendly and impactful. The AI’s ability to suggest diet-specific ingredient alternatives with minimal disruption to the recipe is a feature we’re particularly excited about.
What we learned
Through this project, we gained valuable experience in applying AI to real-world problems, particularly in the area of natural language processing for recipe analysis. We also learned how to integrate different technologies, like MongoDB Atlas and Databricks, into a cohesive app. Most importantly, we learned about the importance of user feedback in refining the user interface and improving the quality of the suggestions made by our AI model.
What's next for Shelf Smart
In the future, we plan to enhance Shelf Smart by adding more dietary filters (e.g., low-sodium, low-sugar) and expanding the recipe database. We also envision a community feature where users can share their modified recipes, helping others discover new meal ideas. Additionally, integrating a meal-planning feature and shopping list generator will make the app a one-stop solution for all kitchen needs.
Log in or sign up for Devpost to join the conversation.