Signal & Noise AWS re:Invent 2025: Beyond the Hype—Agentic AI, Interoperability, and the Return of IRL
What actually mattered at AWS re:Invent 2025—once the keynotes ended and the hype faded?
In this belated, boots-on-the-ground debrief from Las Vegas, Signal & Noise hosts Brett House and Rio Longacre break down the real signals from re:Invent 2025, drawing on executive conversations, daily floor recaps, and firsthand time with partners and practitioners across the ecosystem. Joined throughout by Credera's Alyssa Furth, the episode focuses less on announcements, and more on what’s actually changing inside modern data, marketing, and AI stacks.
Across conversations with partners including Databricks, Tealium, Treasure Data, Claravine, and Jasper, a few themes came through loud and clear:
Agentic AI is real now — moving from buzzword to builder toolkit with services like Nova, Bedrock, Transform, and MCP
Interoperability matters more than ever — as brands struggle to connect fragmented MarTech and AdTech stacks without massive re-platforming
Data foundations are non-negotiable — governance, consent, metadata, and taxonomy are still the gating factors for everything AI promises
IRL is back — re:Invent proved that real relationships and in-person collaboration are accelerating progress faster than months of remote meetings ever could
Rather than chasing shiny objects, this episode surfaces a more grounded truth: AI doesn’t replace fundamentals—it exposes where they’re broken. And the organizations making progress are the ones investing in connective tissue, not just new tools.
If you’re navigating agentic AI, modern data stacks, or the future of marketing and analytics—and want to know what actually changed at re:Invent—this episode separates signal from noise. Enjoy!