Cloud Talk Show with Ralph Rivas and Larry Smithmier - Real-World Cloud AI: RAG, Agents, and Responsible AI ft. Rahul Modi
Ralph and Larry welcome Rahul Modi, a cloud architect who lives and breathes Azure day-to-day — and has the real-world scars to prove it. Fresh from presenting at Louisville AI Week, Rahul walks through how he’s wiring Azure AI Foundry, RAG-based retrieval, multi-agent orchestration, and the AI Gateway together for actual clients in healthcare and government. The conversation gets into why responsible AI isn’t optional when patient data and insurance records are in the mix, how SLMs are quietly replacing LLMs for cost-conscious enterprises, and why Azure’s integrated ecosystem gives cloud engineers a shorter path into AI than most other platforms. Ralph and Larry round it out with a lively debate on what “low code” even means anymore, why your prompts need context more than commands, and the eternal truth that if you don’t give a client a color, you’re going to end up with brown. Links below: Cloud Talk Show on YouTube Azure AI Foundry Azure API Management & AI Gateway Louisville AI Week Microsoft Copilot Ollama (Local LLM Framework) Retrieval-Augmented Generation (RAG) Overview