Schema Generation for Querying a Data Lakehouse

Watch a technical lecture where IBM Senior Research Engineer Balaji Ganesan explores schema generation for data lakehouse querying using Large Language Models (LLMs). Dive into the challenges of generating SQL queries and schemas, examining techniques like schema linking, knowledge infusion, preference learning, and uncertainty quantification. Learn about expanding these concepts to GraphQL for retrieving data from diverse sources including databases and REST APIs. Understand the ongoing research questions around whether existing techniques like in-context learning and fine tuning are sufficient, or if LLMs require additional reasoning capabilities. Benefit from Ganesan's extensive experience in Data and AI, including work with Knowledge Graphs, LLM Agents, Graph Neural Networks, and applications in Semantic Automation.