How Data Teams Use Embeddings to Capture Meaning
from The Data Science Podcast with Fexingo: Analytics, Machine Learning, and Data-Driven Conversations Hosted by Fexingo
In this episode, Lucas and Luna explore how data teams use embeddings to turn text, images, and users into vectors that machines can reason over. They start with the concrete example of Airbnb's 2018 shift to embedding-based search, which boosted bookings by nearly five percent, and walk through why representing meaning as coordinates in high-dimensional space is so powerful. They discuss the trade-offs between static embeddings like Word2Vec and contextual models like BERT, how to evaluate embedding quality beyond cosine similarity, and where embeddings break down—like the time a model encoded 'doctor' as male. They also cover practical concerns: storage costs, GPU requirements, and how to keep embeddings fresh when the world changes. By the end, you'll know how to spot good embedding infrastructure and the warning signs that your semantic search is just a fancy keyword matcher. If this episode gives you something usable, listener support keeps it ad-free—find them at buy me a coffee dot com slash fexingo. #Embeddings #Word2Vec #BERT #SemanticSearch #VectorDatabases #Airbnb #NaturalLanguageProcessing #MachineLearning #DataScience #TechPodcast #FexingoBusiness #BusinessPodcast #DataEngineering #AI #ML #RecommendationSystems #TextMining #HighDimensionalData Keep every episode free: buymeacoffee.com/fexingo
Embed this episode
No transcript yet.