Agentic workflows: Transforming, structuring, enriching data at scale with AI

Every enterprise unfathomable volumes of unstructured data — documents, emails, transcripts, contracts, support tickets, web content — growing faster than anyone can use it. Agentic workflows promise to turn this data into insights and automations, but they're only as good as the data feeding them.AI Functions bridge that gap at scale: research-backed, composable building blocks optimized for batch processing millions of records. From SQL, they parse documents, classify text, extract structured insights, and resolve unstructured data into clean, governed inputs — one pipeline that feeds every downstream consumer. This session shows how to optimize running agents on your data at scale. We'll walk through how to compose AI Functions into end-to-end agentic workflows, and reveal new AI Functions that push batch enrichment and retrieval further than ever before.