🚀 We are launching Anyshift Root Cause Analysis (RCA). Anyshift RCA is the AI on-call engineer that understands every dependency in your cloud infrastructure. It continuously builds a deep knowledge graph from your GCP, AWS, k8s and other platforms, then fuses that context with

Big news for #cloud and #DevOps in Europe (Germany): AWS is doubling down on digital sovereignty with the European Sovereign Cloud: operated *exclusively* by EU citizens, physically located in the EU, and fully independent from non-EU infrastructure. This is a bold answer to.

Google just rolled out a major Gemini AI update: now, your chatbot will remember key details from past convos automatically without you needing to prompt it. 🧠✨.

🛠️ Only 1 in 4 IT teams say their workflows are smooth, a new Temporal survey finds. The rest? Fighting fires. 🔥 Key pain points:.- 35%: Complex workflows & high operational overhead.- 34%: Struggle to recover from failures.- Executives prioritize reliability/security; devs.

'What good GitOps' really looks like?.The latest State of GitOps report highlights the biggest drivers:.- responsive code review .- automated pulls with continuous reconciliation. Automatic reconciliation is especially powerful, yet under-adopted. Teams embracing all 6 core

AWS has launched EC2 G6f instances, introducing fractional GPU support using NVIDIA L4 Tensor Core GPUs. You can now provision as little as 1/8 of a GPU (with 3GB memory), helping teams optimize costs and adapt resources for graphics, engineering, and AI/ML workloads. G6f runs.

What’s new in k0rdent v1.0.0?. Highlights: .- Stabilized APIs. - Multi-cluster support .- GPU and control plane monitoring .- Automated IP management .- Templated service upgrades . Built on open standards with OpenCost, VictoriaMetrics, and Grafana, k0rdent streamlines platform.

🚢 Datadog Integration is live in Anyshift!. When an issue arises, Annie, our AI SRE, directly performs the root cause analysis:.🧭 Detects a slow database query impacting multiple services. 🧭 Traces the issue back to a specific code deploy. 🧭 Pulls relevant Datadog metrics and

Only in SF you get to see such stickers on the street

AI's booming growth reveals even more the importance of data :.- as data storage and compute get cheaper and more efficient, we generate and consume even more data. - but legacy practices still lead many companies to discard valuable data, limiting AI model fidelity and ROI. The.

Software engineering now runs on 2 tracks:.1/ disposable code for speed, and 2/ durable code for reliability. AI accelerates both, but maintainability and trust come from real-world use, automation, and testing. Fast experiments are rising, yet the world relies on robust.

Cool update for EC2 Auto Scaling: you can now trigger AWS Lambda directly from lifecycle hooks. 🌀. This means you can now run custom code during instance state changes, like gathering logs before termination, without EventBridge, SNS, or SQS. ->Fewer dependencies mean.

Time to shift from Human-in-the-Loop (HITL) to Human-on-the-Loop (HOTL)? 🤖. -> HOTL means structuring AI autonomy with real-time observability, least-privilege tooling, triggerable interventions, verification pipelines, and postmortem-ready logging. Read more below 👇.

RT @dexhorthy: couple more highlights from the show

👀👀

1 hour before AGI. ? 🌚.

AWS Lambda now supports response streaming payloads up to 200 MB, a 10x upgrade!!. Stream real-time AI chats, large PDFs, and media files directly from Lambda without S3 workarounds. Amazing update for latency-sensitive and serverless workflows needing faster response times.

Ever wondered how to make Postgres painnnnnnnnfully slow on purpose? (x42000!!). Tweak postgresql.conf by shrinking shared_buffers, ramping up autovacuum, inflating WAL activity, disabling indexes, and forcing single-worker I/O. You can slow Postgres by 42,000 times without.

Migrations are the hidden threat to system reliability. 🔄 Migrating between old and new systems is always complex and risky. Delays increase vulnerability, yet teams rarely prioritize migration work. Incremental transitions can help but often make systems harder to.

Facing AWS IPv4 exhaustion? Learn how to transition your EKS clusters to IPv6 while staying compatible with your existing Istio Service Mesh. -> Cool lesson to deploy multi-cluster setups across single or multiple VPCs, ensuring smooth traffic and service discovery between.