TL;DR: I prepared my AWS Summit Hamburg talk through 6 recorded dry runs, each analyzed by a structured AI skill. The skill transcribes via Amazon Transcribe, counts fillers with word-level timestamps, checks customer references against a checklist, flags pronunciation issues, identifies filler-dense windows, and proposes tighter alternatives. This post shows the complete skill file with every step explained, the real results (um: 80→27, customer refs: 0/6→4/6), and the key discovery: the skill and the talk improved together. Each dry run made both better. Total cost: $3.60 for 6 analyses.

TL;DR: Michael I. Jordan, the man Science magazine named “the world’s most influential computer scientist,” has never considered himself an AI researcher. His thesis: intelligence is not a property of individual systems but emerges from collectives interacting through economic mechanisms. Most AI today is optimization (one agent, one metric). The real world is an equilibrium problem (multiple agents, competing incentives, stable states). The distinction matters for how you architect systems in media, advertising, retail, and publishing. Jordan’s framework does not say “stop building.” It says “stop optimizing in isolation and start designing markets.”

TL;DR: AI agents face the same discovery problem microservices solved with service registries a decade ago, except agents need semantic search, governance workflows, and dynamic capability updates. AWS Agent Registry (preview) implements this pattern as a managed service. I tested it end-to-end: create a registry, register skills, approve them, and discover via natural language queries. The code and gotchas are below.

The Problem We Already Solved Once

In 2012, Netflix open-sourced Eureka [1]. The problem it solved was simple to state and hard to live without: in a microservices architecture, you can’t hardcode where services live. Instances spin up and down. IP addresses change. New capabilities appear. Old ones get deprecated. Without a registry, every service needs to know the exact location of every other service it depends on. That doesn’t scale.