Where the world meets DevOps. Powered by @TechstrongGroup.

Joined February 2011
Include

Only show posts containing:

Exclude

Hide posts containing:

Time range
-
Minimum likes
Karmada has graduated from the CNCF's technology program, giving the federated control plane for multiple Kubernetes clusters a stamp of approval for production use just in time for the AI inferencing work it needs to support. The project turns multiple clusters into a coordinated fleet with centralized placement, failover and multi-cluster autoscaling, and organizations like Bloomberg and Alibaba Cloud already run on it. Read the full article on the graduation and where the project goes next: zpr.io/SgqRqZU39mQR #Kubernetes #CloudNative #Karmada #CNCF
1
105
DevOps.com retweeted
Open-weight models still leave enterprises guessing. Mike Vizard talks with Gregory Kurtzer, Founder and CEO of CIQ, about OpenWALDO. Watch: techstrong.tv/videos/ai-lead… #AI
4
156
AI agent behavior is non-deterministic by definition, so a single prompt change can degrade response quality even when standard metrics show no errors, and without full trace visibility, diagnosing why an agent took an unexpected path becomes guesswork. AWS's new CloudWatch Omni captures every step in a structured timeline with built-in evaluations for correctness, coherence and tool selection, aiming to give teams a unified view regardless of how the agent was built. Read more about how AWS is approaching agent observability: buff.ly/guZrSxm #AWS #Observability #AIAgents
2
2
1
204
Microsoft warns Anthropic's "conscious" AI threatens control. OpenAI logs 6 misalignment cases. Dreamforce: AIforce, Koa, $2B Listen Labs bid. IBM Anderon starts quantum wafer runs. Guests @jswartz Jaime Hampton Jiewen Wang. #AISafety #Dreamforce nitter.cf/i/broadcasts/1pJdRDlBL…
4
1
1
451
Interesting read on why shift left is dead - but is it really? That was a big buzzword in the DevOps/DevSecOps world
DevOps.com
@devopsdotcom
Sep 23
Shift left is dead, and it died because it was fundamentally reactive, waiting for code to exist before it could act. AI-driven development introduced a category of risk it was never built to address, exposure that happens before a single line is generated, the moment a developer connects an untrusted agent, installs a malicious package or exposes a secret in a prompt, and 81% of organizations still lack full visibility into how and where AI is being used across the software development lifecycle. Read why security needs a new model for the agentic development lifecycle: buff.ly/Gg1FhM1 #DevSecOps #AppSec #AIAgents #SoftwareSupplyChain
1
313
AI ambition is easy. Turning it into measurable value is where the real work begins. Join @RedHat’s virtual event, 𝘛𝘩𝘦 𝘈𝘐 𝘍𝘢𝘤𝘵𝘰𝘳𝘺: 𝘍𝘳𝘰𝘮 𝘝𝘪𝘴𝘪𝘰𝘯 𝘵𝘰 𝘝𝘢𝘭𝘶𝘦, for a practical look at how open architecture, accelerated computing and a strong partner ecosystem can support enterprise AI in production. The agenda brings together perspectives from @RedHat, @nvidia, @Dell and @TechstrongGroup on: 🔹 The current state of enterprise AI 🔹 Building an open foundation while keeping technology choices flexible 🔹 The AI Factory in action with Red Hat, NVIDIA and Dell Technologies 🔹 AI sovereignty and maintaining control 🔹 Partner solutions designed for real-world AI workloads 📅 October 15, 2026 🕦 11:30 a.m.–2:30 p.m. SGT 💻 Virtual event See how organizations can connect AI strategy with the infrastructure, platforms and ecosystem needed to deliver measurable outcomes. Register here: bit.ly/3TlaoqX #EnterpriseAI #AIInfrastructure #RedHat #NVIDIA
3
3
154
DevOps.com
@devopsdotcom
Sep 25
Blitzy launched a free sandbox that lets DevOps teams reverse-engineer up to 1M lines of code and use AI agents to identify and remediate vulnerabilities. Read more: buff.ly/efWJqFw #Blitzy #DevOps #DevSecOps #AICoding #SoftwareSecurity
1
665
DevOps.com
@devopsdotcom
Sep 25
AI agents can act with user authority and access sensitive data. Can you prove who authorized what? Explore agent identity, least privilege and audit in this hands-on AWS workshop. 📅 Oct 1 | 3 PM ET 🎟️ Register: buff.ly/jXXK0pr #AISecurity #AWS #CloudSecurity
1
598
DevOps.com
@devopsdotcom
Sep 25
3,200 vulnerabilities fixed. 98% stayed fixed. Zero human security review. See how StackHawk used AI coding agents to find, fix and verify vulnerabilities in production. 📅 October 1 | 1 PM ET 🎟️ Register now: buff.ly/RcKPrS1 #AppSec #AISecurity
1
742
DevOps.com retweeted
Marino Petriccione
@creationloopceo
Sep 23
Replying to @devopsdotcom
Agents that install packages on their own need a gate before the first tool call. creationcode makes you approve a readable plan and set the maximum build cost before the run starts. A human accepts the verified PR before merge. The final receipt closes the case.
1
103
DevOps.com retweeted
Enterprise AI agents do not have a model problem. They have a data access problem. Mike Vizard talks with Tim Bond, Chief Product Officer at Adeptia, about data products, governance and why agents need systems of action. Watch: techstrong.tv/videos/ai-lead… #AI
1
2
1
502
DevOps.com
@devopsdotcom
Sep 25
AI cost management isn’t just Cloud FinOps 2.0. The economics are different. Explore AI cost attribution, governance and waste with Harness’ Patrick Brogan. 📅 October 1 | 9 AM ET 🎟️ Register now: buff.ly/EHwMlVv #AICostManagement #FinOps #AI #CloudFinOps
1
493
Replying to @devopsdotcom
agents installing packages unsupervised is how you inherit yesterday's CVEs at 3am. I pin a allowlist and fail the run on anything else.
1
109
DevOps.com
@devopsdotcom
Sep 24
Talentica Software this week launched DevX AI Pods, a managed software delivery service that uses AI agents trained to supply the context needed to deploy applications built with AI coding tools at scale. The agents evaluate code against a CCCR framework while more than 600 Talentica engineers validate the output, addressing a gap where AI tools generate more code without actually increasing the rate at which applications ship. Read the full article on the service and the survey data behind the code volume problem: buff.ly/exiItJq #AI #DevOps #SoftwareDevelopment
1
516
DevOps.com retweeted
Replying to @devopsdotcom
difference is blast radius. skip-permissions trashes a sandbox, auto-merge ships to everyone. only works if the agent cant touch the gates: ci config, test files, codeowners all locked. otherwise it just learns to make checks green
1
91
DevOps.com retweeted
Behavior drift is the agent security risk most teams won’t see coming. Alan Shimel talks with David Mytton, Founder and CEO of Arcjet, about securing AI agents at runtime. Watch: techstrong.tv/videos/intervi… #AISecurity
3
2
293
the compute access part is the one people forget to plan for until they're already stuck waiting on it
1
90
DevOps.com
@devopsdotcom
Sep 24
Most security teams can tell you how many people work in engineering, but far fewer can tell you how many credentials those people and their installed apps hold against the company's source code. GitHub Enterprise Cloud owners can now export a full inventory of SSH keys, PATs, OAuth tokens and GitHub App tokens with last-used dates, and coding agents are only going to make that count harder to keep up with. Find out what the export covers and why the first one will turn up credentials nobody knew were still active: buff.ly/RjedSLm #GitHub #DevSecOps #Security #SupplyChainSecurity
496
DevOps.com
@devopsdotcom
Sep 24
An OAuth token taken from a former CrowdSec employee's laptop still had permission to read the company's private repositories, and by the time GitHub access was revoked three days later, roughly 170 private repos had already been dumped. The theft went undetected until the source code surfaced on a dark web marketplace in September, and the CEO's own conclusion is that the company was very close to clean. Read the full article on the TeamPCP supply chain attack and the offboarding gap it exposed: buff.ly/ZMmFx7e #Cybersecurity #SupplyChainSecurity #CrowdSec #TeamPCP
1
844