Semantic Recipe Search Using Couchbase and Embeddings

I will present a recipe recommendation application where recipe ingredient lists are converted into vector embeddings using a text embedding model. These embeddings are stored in Couchbase Capella.
When a user enters a list of ingredients they have, the app converts their input into an embedding and performs a vector similarity search to find the closest matching recipes.
The application also includes a web scraper to regularly collect new recipes and ingredients from food websites, embedding them and adding them to the Couchbase database to enrich the dataset over time.