I am Avi, a Solutions Architect at Databricks. We have built an application to demonstrate how AI-capabilities could be easily integrated to deliver novel user experiences. The application allows users to submit images and text, and uses these inputs to return related recipes from a collection of over 13K+ recipes and images scraped from the Epicurious website.
This application consists of a React front-end stylized using Bootstrap and deployed using GitHub pages. An Azure App Service backend calls a REST API presented by Databricks Serverless Real-Time Inference. MLFlow provides several capabilities for packaging models for deployment, which allowed us to easily deploy a pre-trained SentenceTransformers model with a pickled set of recipe data and image embeddings. To help folks understand exactly how this is done, we’re providing access to all the front-end and back-end code behind the app. We hope this helps others explore new ways they might bring AI into their applications.
Check out this super fun blog on how to leverage Databricks model serving to build AI-enabled apps, like our recipe building app. With this app, you can make sure that your turkey is cooked to perfection, your stuffing is delicious, and your pumpkin pie is perfect. This app is the perfect way to make sure that your holiday dinner is a success.
We invite everyone to interact with the application, which we will host throughout the 2022 holiday season. Do let us know your thoughts in the comments 🙂
