Power AI with Real-Time Data using AsyncAPI & Agent Mesh
summary
Realtime change data from Oracle GoldenGate is captured and pushed into AsyncAPI channels, the demo shows GoldenGate creating a data stream and returning the AsyncAPI YAML that is imported into an event mesh; the pipeline preserves source order and offers exactly-once semantics so apps don’t need retry logic. A live airport map displays 20 inserts and 28 updates of flights, people and vehicles, and an agent mesh generates a marketing report from 20 minutes of streaming events via the agent mesh visualizer — the end-to-end flow works live and produces interactive dashboards and campaign recommendations.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 1:09Events for modern apps
- 2:36Agentic AI and orchestration
- 5:03Decoupling agents with event mesh
- 6:26Solace Agent Mesh Framework
- 9:11Airport use case and demo
- 10:35GoldenGate for real-time replication
- 13:06GoldenGate deployment options
- 16:37Change data events via AsyncAPI
- 21:35How GoldenGate data streams work
- 24:25GoldenGate via AI MCP Server
- 26:11Demo setup and real-time capture
- 33:19AsyncAPI with Solace Event Mesh
- 34:21Visualizing airport events
- 38:11Marketing insights with Agent Mesh
- 43:28Conclusion
related talks
26mAI Agents Meet AsyncAPI: Conversational Interfaces For Event Streams.
30mCase Study Productizing AsyncAPI for Data Replication, Changed Data Capture
22marmonizing Asynchronous Systems: NATS, CloudEvents and AsyncAPI.
13mConnecting data sources with AI
31mAsyncAPI is here to stay
40mAI Agents in Practice
07mOwn your Agentic AI
45mFishbowl Discussion on APIs and AI