🤖 AI Model Selection Made Easy: Cisco's Amazon Bedrock Journey | How to Build Like AWS Ep. 4 Ever wondered how to pick the right AI model for your app and actually trust the answers it gives? In this episode, we sit down with the developers behind Cisco's Duo platform to hear how they built a RAG-powered chat assistant on Amazon Bedrock that helps users troubleshoot identity management issues in real time. From benchmarking models on latency to using LLM-as-a-judge for answer validation, this is a real story from engineers who built it, shipped it, and measured it. What we cover: What Cisco Duo is and the problem the team set out to solve How they built a RAG and tool-calling agent system on Amazon Bedrock How they chose the right model — benchmarking, latency testing, and using logs to evaluate LLM performance Why they used Bedrock's model flexibility to go beyond a single provider LLM-as-a-judge and human-in-the-loop: when to use each and why it matters A live demo of the Cisco s...