Power data analytics with generative AI using BigQuery and Gemini

"We are bringing Google’s research and innovations in artificial intelligence (AI) directly to your data in BigQuery. Join this session to learn about BigQuery’s built-in ML capabilities, such as model inferences, and how to use Gemini, Google's most capable and flexbile AI model yet, directly within BigQuery to simplify advanced use cases such as sentiment analysis, entity extraction, and many more.
Speakers: Alicia Williams, Tomas Talius, Seamus Abshere
Watch more:
All sessions from Google Cloud Next → goo.gle/next24
#GoogleCloudNext

[@user-fc6xh8rs4g]

Gemini models in BigQuery - 5:05
GenAI pipelines with Gemini models - 5:58
Use Cases
Text-Based Use Cases - 6:27
Multimodal Use Cases - 7:15
How to use LLMs in BigQuery - 9:40
=> Results : 10:07
Object tables provide a structured representation for unstructured data - 13:23
Vision LM example - 14:15
BigQuery Studio Notebook demo
(move poster images -> title, year extraction (with image model & LM) -> vector embedding -> vector search -> IMDB meta data mapping) - 14:50