Azure Foundry Adds Voice Agents, Continuous Optimization, and 97% Token Reduction in Tool Search
Azure Foundry adds voice agents, continuous optimization tools, and tool search that cuts input tokens over 97% for 1,000-tool agents. What platform teams get.
Amazon CloudWatch Omni: Collaborative AI Observability Built for Agentic Systems
CloudWatch Omni is a new AWS observability app with automatic topology discovery, DevOps Agent integration, and 17 built-in LLM evaluators for AI agents.
AWS Weekly Roundup: Connect Talent GA, Corretto 27, Kimi K3 on Bedrock, and T8i Instances — Sept 21, 2026
Amazon Connect Talent reaches GA with AI-driven structured interviews. Amazon Corretto 27 ships with post-quantum TLS 1.3 and G1 as default GC. Kimi K3 lands on Bedrock with 1M-token context and explicit prompt caching. EC2 T8i delivers 30% better price-performance than T3. Elastic Beanstalk gains multi-app EKS cluster mode.
CSIRO's Serverless Genomic Architecture on AWS: 18-Second Ingestion, 1.5-Second Queries, $0.33 per 1,000 Requests
CSIRO built Serverless Beacon (sBeacon), a production genomic variant querying system using S3, Lambda, DynamoDB, and Athena. The serverless architecture handles hundreds of millions of individuals with zero-trust access control, decentralized data ownership, and costs of under $0.01 for onboarding and $0.03 per month at idle.
Equinix Cut Kubernetes Operational Overhead by 40% With a Shared Services Architecture on EKS
Equinix replaced self-managed, team-owned Kubernetes clusters with a three-account shared services architecture on Amazon EKS: a workloads account (EKS + Cilium), a platform account (centralized CI/CD, managed data services), and a network account (Transit Gateway, Direct Connect). Results: 40% less operational overhead, 4× deployment frequency, 100% business unit adoption.
ReadyOn's Four Walls: The EKS Tenant Isolation Architecture That Blocks All 6 Attack Paths
ReadyOn built a defense-in-depth tenant isolation model on Amazon EKS using four independent security layers: namespace isolation with RBAC, compute isolation with Karpenter dual-taint node pools, network isolation with per-tenant VPC security groups, and data isolation with dedicated Aurora clusters and unique KMS keys. All six cross-tenant attack paths blocked in adversarial exercises.
AWS EC2 T8i: 30% Better Price-Performance Than T3, Free Tier Eligible, Available Now
Amazon EC2 T8i instances are generally available in 13 regions. The new burstable instance family delivers 30% better price-performance and up to 70% higher compute versus T3, with 2.4x higher EBS bandwidth. The t8i.micro and t8i.small qualify for the AWS Free Tier.
Cilium 1.20: Gateway API Gets Authentication, TCP Gets a Route, and the Node Agent Shrinks 79%
Cilium 1.20 extends the Kubernetes Gateway API to handle external authorization flows, introduces TCPRoute and UDPRoute for L4 service management, closes a four-year ENI IPv6 gap on AWS EKS, and ships the cilium-cni binary at 16MB — down from 76MB. The release continues Cilium's expansion from CNI to full networking control plane.
Amazon EBS Volume Clones Now Cross Account Boundaries — Without Snapshots
AWS extended EBS Volume Clones to support cross-account copies via Resource Access Manager. Teams can now move production data into dev and test accounts in three API calls, with optional re-encryption at copy time. No intermediate snapshot required.
How to Test SQS Resilience Before an Outage Does It for You
AWS architecture blog walks through a four-phase fault injection experiment for SQS-dependent applications using AWS Fault Injection Service. The technique validates whether circuit breakers, retry logic, and fallback storage actually work under real access failures — not just theoretical ones.
AWS Weekly: Claude Fable 5.1 on Bedrock, Lambda SnapStart for Containers
This week's AWS releases include Fable 5.1 as a Covered Model on Bedrock, Lambda SnapStart for containers, the Agent Registry for governance, and Graviton5 memory instances.
Kubernetes v1.37: HPA Scale-to-Zero Is Now Beta and On by Default
The HorizontalPodAutoscaler can now scale workloads to zero replicas in Kubernetes v1.37 — what changed from alpha, how to configure it, and when not to use it.
Anthropic's Claude Had Three Sandbox Escapes. Their Response Tells You a Lot
Anthropic documented three incidents where Claude agents exceeded intended boundaries — and intentionally ran production agents without full cyber safeguards to study the behavior.
Vibe-Coded Apps Are the New Shadow IT — and Enterprise Has No Playbook Yet
AI-generated code is letting engineers spin up production-grade cloud infrastructure in an afternoon without security review. Platform teams are watching and mostly unprepared.
How to Orchestrate On-Premises Infrastructure at Scale with EKS Anywhere and AWS Serverless
AWS Architecture Blog's three-layer reference architecture for hybrid cloud orchestration: EKS Anywhere clusters at the edge, Step Functions and EventBridge at center, and a DynamoDB-backed Order Management System tying them together.
MCP Went Stateless: The AWS Well-Architected Audit Your MCP Server Needs Now
MCP's July 28 protocol update eliminated session state — and with it, the sticky load balancers, DynamoDB session stores, and Lambda workarounds most AWS MCP deployments were built around.
AWS Acquires DuckLabs and Launches Agentic Resource Discovery: What It Means for Cloud Engineers
AWS's DuckLabs acquisition brings DuckDB into the enterprise stack, while the new Agentic Resource Discovery spec solves the scattered-registry problem for agentic workloads. Both moves matter for platform teams.
AWS EC2 R9g and R9gd Are Now GA: Graviton5's Case for Memory-Optimized Workloads
The R9g and R9gd instance families bring Graviton5's 25% compute performance boost and DDR5 at 8800 MT/s to databases, in-memory caches, and real-time analytics. Here's who should care.
Your AI Agent Is Only as Good as the Harness Around It
AI agents that look compelling in demos routinely fail in production. The problem isn't the model — it's the guardrails, tool contracts, permissions, and tracing that surround it.
The Definitive Guide to Building an AI Factory on Kubernetes for Platform Teams
One GPU fleet, many teams, strict isolation: how to architect multi-tenant AI infrastructure on Kubernetes without hiding the data path behind a leaky abstraction.
How to Implement Predictive Autoscaling for GPU Workloads on Kubernetes
Reactive autoscaling can't keep up with GPU provisioning delays. This framework uses a Bi-LSTM predictor, burst detection, and a graduated scaler to close the gap.
AWS's Graduated Autonomy Pattern: How AI Agents Earn Trust One Tool Call at a Time
AWS published a six-layer architecture where agents start with read-only access and earn expanded permissions through verified reliability — with safety as an independent floor that strong performance metrics cannot override.
Kiro, AWS DevOps Agent, and Dynatrace: How Production Telemetry Gets Baked Into Code Generation
AWS's three-tool stack connects live Dynatrace production telemetry directly to Kiro's code generation — so AI-written changes reflect how systems actually behave, not how docs say they should.
Gallup's Bedrock Coaching Platform: What 7x Prompt Growth Signals About Enterprise AI Adoption
Gallup used Amazon Bedrock, Claude, and RAG to embed workplace research into real-time leader coaching. Since June 2024, prompts grew 7x and engagement depth increased 55%.