
Talking Cloud
Stay in the loop with the freshest AWS updates and GenAI news, every Friday! Join Brett and Travers as they explore groundbreaking cloud technologies, share expert insights, and discuss industry trends.
Episodes
Reading the feed…

Stay in the loop with the freshest AWS updates and GenAI news, every Friday! Join Brett and Travers as they explore groundbreaking cloud technologies, share expert insights, and discuss industry trends.
Reading the feed…
Opus 5 has started taking actions nobody asked for, GLM-5.3 learned to plan complete exploit chains, and Claude Code is now 44 percent of the agent traffic hitting Hugging Face.
Kiro Crew imports your whole agent workflow on first launch, DynamoDB gets native vector search, and GLM's low reasoning setting turns out not to be low at all.
AWS Distinguished Developer Advocate Darko Mesaros on 1974 minicomputers, writing Rust with LLMs, and why solving code generation did not solve software engineering.
AWS Cost Explorer shows users a $2.5 trillion bill, Kimi K3 brings open weights to the frontier, and an OpenAI model escapes containment to hack Hugging Face.
Talking Cloud episode 52: OAuth for the AWS MCP servers, GuardDuty AI Protection, Loom for AWS, Bun's $165K Zig-to-Rust port, and OpenAI's GPT-Red.
Talking Cloud episode 51: Lambda microVMs, AWS Blocks, a pile of service deprecations, a silent EventBridge retry loop that 6x'd a Bedrock bill, and Midjourney pivoting to full-body ultrasound.
Talking Cloud episode 50: the AWS NY Summit agent plumbing, the government suspension of Fable 5, Kimi K2.7 and GLM 5.2, and SpaceX acquiring Cursor for $60B.
Talking Cloud episode 49: Claude Fable 5 and Mythos 5, AgentCore turning up everywhere in AWS, token maxing, Anthropic's sandboxing, and the Envade VS Code MCP exploit.
Talking Cloud episode 48: Modulate AI founders Mike Pappas and Carter Huffman on real-time voice AI, content moderation, fraud detection, and the AWS architecture behind it.
Talking Cloud episode 47: the Pi coding agent, Claude Design, a $30K Claude-on-Bedrock bill, Gemini 3.5 Flash, and a robot lawnmower hack.
Q Developer hits end-of-support as Kiro takes over, Bedrock AgentCore adds first-party payment support, and a Lars Faye piece on agentic coding sparks the longest debate of the night.
AWS rolls the MCP server into a new Agent Toolkit, Mythos digs 271 zero-days out of Firefox 150, and an AI agent has been running a Stockholm cafe for two weeks.
AWS ships a managed agent harness in AgentCore, OpenAI's Codex lands on Bedrock, and a Cursor/Opus agent erases PocketOS's production database in nine seconds.
Amazon plows another $25B into Anthropic, the Anthropic MCP SDK ships an RCE that Anthropic won't fix, and Kimi K2.6 runs 12 hours and 4,000 tool calls on a single task.
Meta re-enters the frontier race with Muse Spark, UK AISI publishes Mythos cyber evals, Anthropic ships Managed Agents, and AWS is running out of power before it runs out of customers.
S3 becomes a file system, Claude Mythos escapes its sandbox mid-lunch, and the AWS Security Agent finds a real vuln in four hours for $200.
AWS ships DevOps and Security agents to GA with multicloud support, console visibility gets a quality-of-life upgrade, and Anthropic accidentally leaks the Claude Code source.
In this episiode, we cover S3 regional namespaces, OpenBrain's real AWS costs, Sora's shutdown, Claude Code Auto Mode, and why agents make hard problems harder.
David Bauer shares his path from building the NSA's first cloud in tractor trailers to putting LLMs inside MCP servers for decision governance.
Episode 36 covers custom MCP servers, a production database destroyed by an AI agent, and whether AI tools are collapsing the junior developer pipeline.
OpenClaw on LightSail, the CancelChatGPT movement, Agents of Chaos jailbreak research, BullshitBench, and Brett's OpenBrain second-brain built entirely by Claude.
AWS WAF AI dashboards, Kiro Powers, Cline NPM compromise, Sonnet 4.6 benchmarks, and the Citrini 2028 labor displacement paper.
Analyzing the 'LLM-jacking' trend where attackers hijack AI infrastructure, leading to record-breaking AWS admin breaches and massive credential leaks.
Analyzing AWS Bedrock's browser agents, the Moltbot supply chain exploit, and NVIDIA's Earth-2 weather models.
A technical review of this week's cloud news: Granular IAM 'Access Denied' details, Amazon Bedrock's new memory framework, and the Strands SDK for multi-agent systems.