AI is everywhere in cloud security right now. Nearly every product claims to be “AI-powered,” and copilots and chatbots promise to help teams interpret issues faster. But for most platform teams, understanding the problem isn’t the hard part.The real challenge is resolution.Cloud environments change quickly, and the backlog of security findings grows just as fast. Teams aren’t struggling with awareness; they’re buried in alerts, tickets, and rework. What they need isn’t more explanation. They need help fixing things.That’s where most AI tools fall short. They summarize and suggest, but rarely act. What’s missing is AI that understands infrastructure, aligns with…
Author: drweb
This month’s TSQL Tuesday invite is from my good friend, long standing MVP and community volunteer Taiob Ali – Taiob’s call is to blog on how AI, (the biggest invention since the internet, according to some) is changing our careers.The place I work is passionate about AI adoption. We are exploring many tools in that regard. I may not be able to share details of exact usage for privacy reasons. These are my personal experiences.How I use it personallyI have not played with many AI tools. I use a paid version of ChatGPT, which I find helpful for the following…
When I started my journey in DevOps nearly a decade ago, things were very different. CI/CD was barely catching on in enterprise environments, and Kubernetes felt like something only a few elite engineers could master. Over the years, I’ve worked across different kinds of organizations — startups, mid-sized teams and large-scale enterprises — and realized that many DevOps challenges are universal. Whether it’s wrangling outdated infrastructure, optimizing deployments or improving system observability, the fundamentals don’t change. Here are a few lessons I’ve picked up along the way — some from success, many from hard-earned failures. 1. Expect Failures — and Prepare for Them Proactively I…
A survey of 300 mobile engineers in the U.S and the United Kingdom (UK) working for organizations with 500 to 10,000 employees finds, on average, respondents are spending five hours per release on low-value tasks involving manual steps, coordination issues and approval bottlenecks, translating into 130 wasted engineering hours annually per developer.Published by Runway, a provider of a software-as-a-service (SaaS) platform for managing mobile application releases, the survey finds more than three-quarters (77%) said their teams experience incidents requiring delays or hotfixes every three to five releases. Most teams (86%) spend a significant amount of time firefighting instead of building…
Building AI agents can be a complex task. But it also can be a fairly simple combination of answers to the following questions: What is the AI backend that powers my intelligent fuzzy computation? What tools do you need to give to the AI to access external systems or execute predefined software commands? What is the application that wraps these together and provides the business logic for the agent (like when you’re building a marketing agent, what makes it know more about marketing or your particular use-cases than a generic chat-GPT model)? A very popular way to build agents currently…
We’re in the middle of a developer renaissance powered by AI — yet a large portion of the community remains cautious, especially when it comes to the next frontier: Agentic AI. The just‑released 2025 Stack Overflow Developer Survey paints this nuanced portrait vividly: 84% of respondents use or plan to use AI tools in their workflows, yet less than a third trust their accuracy — just 29% this year, down from much higher levels in prior surveys.A Tale of Two AI Worlds: Generative vs. AgenticOn one hand, generative AI tools — ChatGPT, GitHub Copilot and others — have become near-ubiquitous.…
Minimus has extended its managed service for providing application developers with hardened images to include support for the Vulnerability Exploittability eXchange (VEX) format used to share data across multiple application security tools and platforms along with hardened Helm charts for securely deploying applications on Kubernetes clusters.Additionally, Minimus has added compliance dashboards and views, and integration with Microsoft for Single Sign-On (SSO) service.Minimus CTO John Morello said the overall goal is not just to provide access to a set of hardened images but also make it simpler for DevSecOps teams to operationalize them.For example, Minimus now provides hardened Helm charts aligned…
Thomas Dohmke, CEO of Microsoft Corp.-owned GitHub, said on Monday he’s stepping down and will leave the company at the end of 2025 to become “a founder again,” signaling a major shakeup in how the artificial intelligence (AI) coding unit is run.Since Microsoft acquired GitHub in 2018 for $7.5 billion, it has operated as an independent unit. But the exit of Dohmke, who was CEO for nearly four years, marks a major turning point in the way GitHub operates. Microsoft isn’t naming a new CEO position, and the rest of GitHub’s leadership team will report directly to Microsoft’s new CoreAI…
In a tech landscape dominated by distributed systems, serverless architectures, and real-time analytics, one might assume that SQL, a language born in the 1970s, would be fading into obscurity. Yet, SQL continues to thrive, evolving alongside cloud-native databases and remaining the backbone of modern data operations.The Enduring Appeal of SQLIn a world where data pulses beneath every digital surface, one language continues to thread its way through the veins of enterprise logic and analytical precision: SQL. Not because it’s trendy, but because it’s timeless. SQL isn’t just a tool; it’s the grammar of structure, the syntax of understanding, the quiet…
There’s a weird contradiction in modern AI development. We have better tools than ever. We’re building smarter systems with cleaner abstractions. And yet, every time you try to swap out a component in your stack, things fall apart. Again. This isn’t just an inconvenience. It’s become the norm. You’d think with all the frameworks and libraries out there (LangChain, Hugging Face, MLflow, Airflow) we’d be past this by now. These tools were supposed to make our workflows modular and composable. Swap an embedding model? No problem. Try a new vector store? Easy. Switch from OpenAI to an open-source LLM? Go…