Everyone wants an AI assistant. No one wants to maintain the infrastructure behind one. In this practical session, Rakia Finley unpacks the cloud-native realities of building AI agents that aren’t just cool but stable, scalable, and secure.

You’ll learn what RAG workflows actually demand of your storage and container orchestration, how to build latency-aware event pipelines that play nice with LLMs, and what monitoring looks like when your system is generating outputs you can’t always predict. Whether you’re prototyping GenAI features or scaling a production-ready AI tool, this talk helps you make sure your stack is ready for the complexity ahead.