Innovative leaders such as NielsenIQ are increasingly turning to a data lakehouse approach to power their Generative AI initiatives amidst rising vector database costs. Join us for a technical deep dive into the pivotal role of vector embeddings in AI and a demo of how you can generate and manage vector embeddings with the cost and scale efficiencies of your lakehouse.
What You Will Learn:
Real-World Applications: In this talk, we’ll cover the challenges of generating, storing, and retrieving high-dimensional embeddings, including high computational costs and scalability issues for production workloads. Kaushik Muniandi, engineering manager at NielsenIQ, will explain how he leveraged a data lakehouse to overcome these challenges for a text-based search application, and the performance improvements he measured.
Introduction to AI Vector Embedding Generation Transformer: Discover how Onehouse solves the above challenges by enabling users to automatically create and manage vector embeddings from near real-time data ingestion streams to lakehouse tables without adding complex setups and extra tools.
Technical Deep Dive: Get into the nitty-gritty of Onehouse stream captures and how they integrate with leading vector databases, enabling a single source of truth for AI model training, inference, and serving.
Can't make it? Register anyway to receive the recording!
*This post is created by Onehouse. We thank the Onehouse team for their insights and ongoing support of TheSequence.
#Ai, #AiModel, #Applications, #Approach, #Data, #DataIngestion, #DataLakehouse, #Database, #Databases, #Embeddings, #Engineering, #Generative, #GenerativeAi, #How, #Inference, #Insights, #Issues, #It, #Learn, #Model, #ModelTraining, #Performance, #Power, #Production, #RealTime, #RealTimeData, #Recording, #Scale, #Search, #Storage, #Streams, #Tables, #Text, #Time, #Tools, #Training, #Transformer, #Vector, #Webinar
Published on The Digital Insider at https://is.gd/mHvET0.
Comments
Post a Comment
Comments are moderated.