News and Tech Update

I’m going to say something that will make every engineering manager uncomfortable: Stop asking your team to write documentation . Not because documentation doesn’t matter. It matters more than ever. But because asking humans to document their work after they’ve done it is a process that has failed consistently for thirty years, and no amount of “definition of done” checklists or documentation sprints is going to fix it. The people who know the most write the least. The docs that get written are stale within weeks. And the knowledge that matters most — the decisions, the gotchas, the “why” behind the code — rarely makes it into a document because it’s not the kind of thing you sit down and write. The Documentation Death Spiral I’ve watched this cycle play out on every team I’ve been part of: Week 1: “We need to document this.” Everyone agrees. Someone creates a Confluence space. Week 4: A few pages exist. They’re pretty good. Written by the one person who cares about docs. Week...

LocalStack today announced it has extended its ability to simulate Amazon Web Services (AWS) environments to provide an ability to debug applications before deploying them. Company CEO Colin Neagle said App Inspector makes it possible for developers to debug their applications running in a simulated AWS environment inside a container on a local server. Simulating the full application stack within a local sandbox container makes it possible to better understand application behavior such as data flows between AWS services, event execution paths and resource dependencies that may have been inadvertently misconfigured, noted Neagle. Once discovered, App Inspector then generates a visual representation of the interaction between services in the local environment to make it simpler to debug applications without digging through logs and then needing to upload a fix to a staging server running in the AWS cloud. That capability doesn’t replace the need for an observability platform but i...

The era of the “human-only” software engineer is rapidly receding into the rearview mirror. Google CEO Sundar Pichai revealed Wednesday that a whopping 75% of the company’s new code is now generated by artificial intelligence (AI), marking a major shift in how the tech giant builds its products. The velocity of this transition has caught even industry observers off guard. Just 18 months ago, in early 2024, AI-generated code accounted for only a quarter of Google’s output. By late 2025, that figure had climbed to 50%. Today’s 75% milestone signals that AI has moved from a supplemental “autocomplete” tool to the primary engine of production at Google. Pichai noted that the workflow has evolved into something “truly agentic.” Instead of human engineers laboriously writing lines of code or using AI to finish a single sentence, they are now supervising autonomous digital teams. These AI agents can plan, execute, and refactor entire codebases with minimal human intervention. The effic...

Did God vibe the universe into existence? My mind served up a strange thought at three in the morning. The sudden idea may have been sparked from an occurrence in the previous evening, when one of our team members spun an entire e-commerce website by merely “vibing ”  with Codex. I tried to shush my mind, but it wouldn’t stay quiet.    God spoke, let there be light and there was light, isn’t that a classic example of spinning the whole universe by sheer vibing? Now for the record, my mind has never contested or undermined the Big Bang theory, but creating the world through mere words feels far less unbelievable when seen through the  vibe coding analogy . The mind prodded further.    Could God have created and then deputed (abandoned?) the world to human agents, eerily similar to how humans have deputed (are deputing) software development to AI agents?    Possible, entirely possible! Now my eyes were wi...

You can’t make this crap up. You just wish you could. Jer Crane, founder of the small vertical software company, PocketOS , reported on X that the AI Cursor coding agent and a Railway backup misconfiguration combined to briefly wipe out the company’s car‑rental customer production data . Not some of the data. All of it. That’s a company killer. Fortunately for PocketOS and its customers, Crane later reported that Railway had managed to “recover the data (thank God!).” Thanks to that miracle save of reconstructing the missing data from earlier backups, PocketOS and its customers are back in business. But how could this happen in the first place? According to Crane, it was a chain of failures from both Cursor , the AI development environment, and Railway , his infrastructure provider. Together, they created a “perfect storm” that turned a routine staging bug fix into a company‑threatening outage. In his post, Crane recounted how an autonomous AI coding agent running inside Cursor, ...

AI coding agents are becoming more capable, but evaluating them is harder than it looks. Most benchmarks focus on a single dimension of agent capabilities; for instance, the popular SWE-Bench benchmark only focuses on fixing issues on open source Python repositories. Real-world software engineering involves fixing bugs of course, but it is a lot more multifaceted: in any single week a software developer may also debug complex issues, building a new greenfield script or app, improving test coverage, fix bugs on a frontend repo, research unfamiliar APIs – the list goes on. The OpenHands Index addresses this by building a much broader benchmark evaluating language models across five distinct categories: Issue Resolution (fixing bugs), Greenfield development (building new applications), Frontend development (UI tasks requiring visual understanding), Testing (generating tests to reproduce bugs), and Information Gathering (research and documentation tasks). This diversity matters because...

It appears that GitHub has its hands full adjusting to the demands of scaling AI workloads. First, the company paused sign-ups for its Copilot subscription tiers in response to a wave of demand from agentic AI projects. Then it shifted to usage-based pricing to, again, better align revenue with the heavy compute demands of AI projects. Now GitHub is confronting still more infrastructure challenges as it deals with the rapid growth in AI-driven software development. Two recent service disruptions have highlighted the pressure, prompting the company to upgrade its platform for higher capacity and resilience. Tenfold Capacity Boost Is Not Enough GitHub had initially planned for a tenfold increase in capacity beginning in late 2025. Within months, even that ambitious projection proved insufficient. The company is now engineering for a thirtyfold expansion, reflecting both the speed and magnitude of demand tied to AI-assisted development workflows. The urgency, as detailed by GitHu...

OpenAI has unveiled Symphony, an open-source specification that shifts how software development teams deploy AI in workflows, moving from interactive coding assistance toward continuous orchestration of autonomous agents. Symphony reframes project management tools as operational hubs for AI-driven coding. Rather than prompting an assistant for individual tasks, developers assign work through issue trackers, allowing agents to execute tasks in parallel and deliver outputs for human review. The change reflects a trend in enterprise AI in which systems are increasingly embedded into production pipelines rather than used as standalone tools. Symphony emerged from internal experimentation at   OpenAI , where engineers attempted to scale the use of   Codex   across multiple concurrent sessions. While the agents proved capable, human operators became the limiting factor. Engineers found they could only manage a handful of sessions before coordination overhead offset pro...

AI has already arrived in the software development lifecycle; not as a pilot program or controlled experiment, but as an everyday reality. Developers are using AI coding assistants to generate functions, refactor modules, review pull requests, and accelerate delivery, often in direct tension with corporate policies meant to limit or control that use.   While it’s tempting to consider this some kind of ‘Shadow AI’ or ‘Governance Failure’, it is a signal of things to come in this brave new world of AI-accelerated software engineering.   Recent industry surveys show that  well over half of developers now rely on AI coding assistants in their daily work, with many using them frequently or constantly. At the same time, more than three-quarters of organisations have formal policies that restrict or prohibit that same usage. From a security perspective, that tension is understandable but may be misplaced, because from the standpoint of application risk, ...

The development community saw this one coming: GitHub will transition its Copilot service to a usage-based billing model on June 1, replacing its existing system of fixed subscriptions supplemented by premium request limits. As reported last week, GitHub suspended new sign-ups for several of its Copilot subscription tiers as it faced a surge in demand from agentic coding workflows. To address that, under GitHub’s new pricing model, customers across individual, business, and enterprise tiers will receive a monthly allocation of AI credits, which are consumed based on token usage. This includes input, output, and cached data processed by underlying models. Once those credits are exhausted, users can purchase additional capacity at published rates. The change leaves base subscription prices intact. Individual plans remain priced at $10 per month for Pro and $39 for Pro+, while business and enterprise tiers continue at $19 and $39 per user per month, respectively. Each plan’s monthly ...

The Model Context Protocol (MCP) is moving faster than the developer community can keep up with, racing past its original design parameters and leaving teams scrambling to build clients that can match its pace. The result is an ecosystem where the protocol itself keeps shifting under everyone’s feet, and where the tooling, conventions and security thinking that should accompany a foundational standard are still being figured out on the fly. Joey Stout, solutions architect at Spacelift, joins Mike Vizard to make the case that this is the price of being early. Stout describes an environment that increasingly resembles a Wild West, where rogue MCP servers get spun up inside organizations without anyone in leadership knowing they exist, let alone whether they have basic guardrails wrapped around them. The convenience of standing one up in a few minutes has outrun the discipline needed to govern them. MCP servers can give AI agents broad reach into internal systems, data and APIs, an...

DevOps.com is now providing a weekly DevOps jobs report through which opportunities for DevOps professionals will be highlighted as part of an effort to better serve our audience. Our goal in these challenging economic times is to make it just that much easier for DevOps professionals to advance their careers. Of course, the pool of available DevOps talent is still relatively constrained, so when one DevOps professional takes on a new role, it tends to create opportunities for others. The ten job postings shared this week are selected based on the company looking to hire, the vertical industry segment and naturally, the pay scale being offered. We’re also committed to providing additional insights into the state of the DevOps job market. In the meantime, for your consideration. Dice Booz Allen Hamilton Beavercreek, OH DevOps Engineer $61,900 to $141,000 CGI Reston, VA DevOps Engineer $108,300 to $137,100 Noblis Philadelphia, PA DevOps Engineer $86,800 to $135,625 Le...

Building AI agents sounds straightforward until you actually do it. You need an agent to onboard a new employee. It has to create an Entra ID account, provision GitHub access, spin up cloud resources, create tasks in Azure DevOps, and send a welcome message in Teams. Five tools. Five different authentication models. Five different teams are managing those tools. Now multiply that across every agent your organization is building. That’s the problem Microsoft is addressing with Toolboxes in Foundry, now available in public preview. What Toolboxes Actually Do A Toolbox is a named, reusable bundle of tools managed in Microsoft Foundry. You define your tools once, configure authentication centrally, and expose everything through a single MCP-compatible endpoint. Any agent that can consume an MCP endpoint can use a Toolbox — regardless of the framework it was built on. The endpoint looks like this: https://zava.services.ai.azure.com/api/projects/<project>/toolbox/<toolbox-na...

Microsoft has unveiled plans to incorporate Anthropic’s Claude Mythos Preview model and other AI models into its Security Development Lifecycle, embedding AI directly into the stages where code is written and tested. Rather than relying primarily on static analysis tools, Microsoft is adopting AI models capable of analyzing code dynamically and identifying complex vulnerabilities that might otherwise go undetected until later stages of development. Released on April 7, Anthropic’s Mythos model has already demonstrated a previously unmatched ability to uncover critical flaws across operating systems and widely used software. Anthropic claimed that the model’s ability to find security vulnerabilities is so advanced that it should not be released to the public. Microsoft gained access to the model through Anthropic’s Project Glasswing, a program that grants limited access to select tech firms for cybersecurity research. Within this framework, Microsoft is reporting measurable improve...

Most B2B applications collect incomplete data by design. A lead form captures a name and company. A recruiting tool surfaces a LinkedIn profile. An event registration system logs an email address and job title. The record enters your system and sits there, half-formed, waiting for someone to manually fill in the gaps before it can be acted on. This is an architectural problem, not a workflow problem, and solving it at the architecture layer is what separates applications that create operational leverage from ones that just digitize manual work. Understanding how to build contact enrichment into your application using professional data APIs changes  how you think about the data ingestion layer entirely. Rather than passing incomplete records downstream and hoping someone fills in the blanks, you enrich at the point of entry, automatically, before the record ever reaches a human. The Architecture Problem Behind Incomplete Lead Records The gap between the data a user submits and t...