re:Invent 2025 re:cap Amazon Kiro, Nova Forge, Lambda Durable Functions

Cloud Chronicles Podcast EP04: the first government-ordered shutdown of a frontier model as Anthropic's Claude Fable 5 (and Mythos 5) are restricted worldwide on national-security grounds, what that means for AI sovereignty and Europe's open-weight push, the wave of models rushing to fill the vacuum with OpenAI's rumored GPT-5.6, Zhipu AI's GLM 5.2, Kimi 2.7, NVIDIA's Nemotron and Mistral's Devstral 2, a walk through the Artificial Analysis intelligence-vs-price-vs-latency leaderboard. SpaceX's record $60B acquisition of Cursor and the Grok play on Amazon Bedrock. The rise of meta-harness architectures and the agent sandbox boom across AWS AgentCore, OpenAI Codex and Azure MXC, and a first look at why agents need a semantic layer and AWS Context.0:00 Intro & Headlines1:14 Claude Fable 5 – Launch, Secret Rerouting & Government Shutdown4:07 What the Shutdown Means for AI Sovereignty7:57 Filling the Vacuum – A New Wave of Models8:41 Model Rundown – GPT-5.6, GLM 5.2, Kimi 2.7, Nemotron & Devstral 215:49 Artificial Analysis Leaderboard – Intelligence, Price & Latency18:53 SpaceX Buys Cursor for $60B & the Grok Play23:38 Meta-Harness Architecture & the Sandbox Boom26:15 Why Agents Need Context – Semantic Layer & AWS Context29:27 Wrap-Up & Next Episode

Cloud Chronicles Podcast EP03: Anthropic's Claude Opus 4.7 and the cybersecurity-focused Claude Mythos gated preview, Claude Code's auto-permissions and Routines, Claude Design out of Anthropic Labs, Google's open-weight multimodal Gemma 4, NVIDIA's NemoClaw enterprise runtime for OpenClaw, OpenAI's reworked Agents SDK with file-system memory and harness/sandbox separation, the OpenAI Frontier agent control plane, Codex computer and browser use, AWS's GA of the DevOps and Security frontier agents, and Stateful MCP on Agent Core.0:00 Intro1:10 Episode 3 Kickoff1:27 Anthropic Claude Opus 4.76:26 Claude Code – Auto Permissions & Routines11:30 Claude Design – Anthropic Labs14:10 Claude Mephos – Cybersecurity Preview18:41 Gemma 4 – Google's Multimodal Open-Weight Family22:24 Token Economy & Agent TCO25:48 NVIDIA NemoClaw – Enterprise OpenClaw30:51 OpenAI Agents SDK – Harness, Sandbox & File-System Memory38:14 OpenAI Frontier – Agent Control Plane40:35 Codex – Computer Use, Browser & SSH41:34 AI DLC – How the Dev Lifecycle Shifts43:07 AWS Frontier Agents GA – DevOps & Security46:16 Stateful MCP on Agent Core49:34 Wrap-Up & Next Episode

Second episode of the Cloud Chronicles Podcast: we break down the biggest AI and cloud moves of the past weeks. The launch of the AWS European Sovereign Cloud, Apple’s agentic coding in Xcode 26.3, OpenAI’s new Codex desktop app, NVIDIA’s Rubin AI platform, and breakout models like Kimi 2.5 and Gemma 3.0:00 Intro0:36 Episode 2 Kickoff0:53 AWS European Sovereign Cloud (ESC)6:25 Apple Xcode 26.312:29 Codex & Cloud Code – Autonomous Coding Tools18:25 Kimi 2.5 & Gemini Flash – Model Updates29:28 OpenAI Scaling & Database Insights32:19 Nvidia Rubin AI Platform35:20 Gemma 3 – Document & Image Processing38:14 Wrap-Up & Next Episode

Deep dive into AWS re:Invent 2025's biggest announcements. We break down Amazon's AI Factories for on-premises AI infrastructure, the expanded Nova model family including Nova Forge for custom frontier model training, Nova Act for browser automation, and Nova 2 Sonic for speech-to-speech AI. Plus: Lambda Durable Functions and what it means for serverless workflows. Technical analysis for cloud practitioners.