Transforming user experiences with real-time AI Copilots
summary
Live demo shows a real-time flight Copilot called Maya built with Confluent Cloud streaming into Azure services: Azure OpenAI (function-calling LLMs v0613/v1106), Azure Functions, CosmosDB and Azure Cognitive Search. The pipeline ingests events (weather, policy), creates embeddings in real time, reindexes unstructured policy docs, and executes transactional rebookings; an example policy fee updates from $150 to $250 and a weather-delay event triggers an automated rebooking flow. Architecture runs on 30,000+ Confluent clusters and uses Kafka triggers to invoke APIs, and the demo completes the actions end-to-end.
AI-written summary · about
chapters (6)
- 0:00Real-time AI Copilots intro
- 2:21Why real-time data matters
- 3:03Intelligent function calling
- 8:22Streaming AI architecture
- 11:13Demo: flight Copilot assistant
- 18:30Real-time data updates
related talks
45mThe era of the AI Copilot
10mAI Co-Pilot Demo
31mBuild AI-assisted communication workflows for customer engagement
58mNext generation AI for developers with the Cloud
15mAI Studio - Creating and scaling your custom copilots
27mEmpowering Innovation through DevOps & AI Integration
45mWelcome to the world of SQL Copilots
56mBuild your own CoPilot