Jamelle Brown, First Orion | Databricks Data + AI Summit 2023
summary
First Orion processes about 100 billion calls annually, turning at least three packets per call into roughly 300 billion records, fingerprints each call, and classifies scams, telemarketing or banking calls while it must respond within 100 milliseconds. Production stack runs 100% on Databricks; experiments include Dall‑E for an internal knowledge base (Maestro) that reports ~7.5x productivity and about $900 savings per employee, plus ML for scam detection and vetting.
AI-written summary · about
chapters (5)
- 0:00Introduction to First Orion
- 0:44Analyzing Billions of Calls
- 3:22AI for Internal Efficiency
- 6:20Leveraging Databricks Tech
- 9:22Future of Data Privacy
related talks
19mDatabricks
24mJoseph K. Bradley, Software Engineer, Databricks Inc. @ MLconf SEA
23mAnalystANGLE
16mTransforming Analytics With AI: A Comprehensive Guide to an Advanced Tech Stack
34mOverview of the Databricks Platform
02h 52mKeynote Day 1 - Full
18mDatabricks drives predictive maintenance, grid optimization with analytics
49mGenAI for Business on the Modern Data Stack
