-
-
web ui
-
aws rag
-
architecture
-
gen ai
-
aws bedrock
CulinaryAI: Smart Recipe Assistant
Inspiration The inspiration behind CulinaryAI stems from the desire to make cooking accessible, fun, and personalized for everyone. We wanted to leverage advanced AI technologies to help users discover new recipes, make the most of available ingredients, and enhance their cooking skills.
What it does
CulinaryAI is a smart recipe assistant that suggests recipes based on the ingredients users have on hand, generates images of the dishes, and provides detailed cooking instructions. It also offers personalized recommendations, ingredient substitutions, and nutritional analysis.
How we built it
We integrated Amazon Bedrock's generative AI for recipe suggestions and image generation. LaunchDarkly's feature flags were used for controlled feature rollouts and experimentation. Convex AI was employed for ingredient substitutions, nutritional analysis, and optimizing cooking times. The app was built using Flask and various AWS services to ensure scalability and reliability.
Challenges we ran into
One of the main challenges was training the AI models with diverse and high-quality data to ensure accurate and relevant recipe suggestions. Integrating multiple AI services and managing feature rollouts with LaunchDarkly also required careful planning and execution.
Accomplishments that we're proud of We successfully created an app that combines cutting-edge AI technologies to provide a seamless and enjoyable user experience. The ability to personalize recipe recommendations and generate realistic dish images are significant achievements that enhance user engagement.
What we learned
We learned the importance of high-quality data in training AI models and the value of user feedback in refining app functionalities. Integrating feature flags allowed us to experiment with new features safely and iteratively improve the app based on real user insights.
What's next for CulinaryAI: Smart Recipe Assistant Next, we plan to add more advanced features such as voice interaction, real-time cooking assistance, and integration with smart kitchen appliances. We also aim to expand our recipe database and further enhance personalization to cater to a wider range of dietary preferences and restrictions.
Log in or sign up for Devpost to join the conversation.