<h2>Must know</h2>
<ul>
<li><strong>AI coding tools are changing enterprise IT hiring and cloud costs</strong>: India’s top IT outsourcers (HCL, Infosys, TCS, Wipro) have drastically cut general hiring, focusing instead on AI-specific skills, despite continued revenue growth. This shift, reported by multiple sources, indicates AI is reshaping workforce needs. Concurrently, Microsoft’s Copilot AI feature is consuming significant capacity units on Fabric platforms, leading to unexpected cloud resource costs for users. This highlights that while AI drives efficiency, it also introduces new operational expenses and demands a strategic realignment of IT resources. <a href="https://app.daily.dev/feed-by-ids?id=DLuyoqNjG&amp;id=nKPk64Gpl" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GitLab 18.8 launches AI-orchestrated DevOps platform</strong>: GitLab 18.8 introduces the Duo Agent Platform, enabling AI agents to orchestrate tasks across the software development lifecycle. This includes a Planner Agent for managing work items and a Security Analyst Agent that uses natural language processing to triage vulnerabilities, even automatically dismissing irrelevant ones. The platform also features a Central AI Catalog for agent discovery and management, aiming to automate security and planning within DevOps pipelines, according to GitLab’s official release. <a href="https://app.daily.dev/posts/R9WyVAVPv" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AMD Zen CPUs vulnerable to StackWarp attack</strong>: A new hardware vulnerability, StackWarp (CVE-2025-29943), affects AMD Zen 1–5 CPUs, including EPYC 7003, 8004, 9004, and 9005 series. This flaw allows privileged attackers to exploit an undocumented control bit to manipulate the CPU stack engine, corrupting the stack pointer and hijacking control flow in SEV-SNP-protected confidential virtual machines. While AMD has released microcode updates, additional firmware patches are expected, and disabling hyperthreading on critical systems is advised as a mitigation. <a href="https://app.daily.dev/posts/9DURDnZX3" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Microsoft open-sources XAML Studio under .NET Foundation</strong>: Microsoft has open-sourced XAML Studio, a lightweight prototyping tool for WinUI and UWP projects, under the .NET Foundation, with version 2.0 featuring a redesigned UI and deeper WinUI 3 integration. <a href="https://app.daily.dev/posts/hUJrXK08w" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Malicious Chrome extensions hijack enterprise sessions</strong>: Five Chrome extensions, disguised as productivity tools, were found stealing authentication session tokens from enterprise HR and ERP platforms like Workday and NetSuite, bypassing MFA and allowing full account takeover for over 2,300 users. <a href="https://app.daily.dev/posts/3ImoRkfEG" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Sequoia Capital invests in Anthropic despite rival stakes</strong>: Sequoia Capital is investing in Anthropic’s $25 billion funding round, valuing the company at $350 billion, despite already holding stakes in competitors OpenAI and xAI, a move that breaks traditional VC norms against backing competing portfolio companies. <a href="https://app.daily.dev/posts/yuiHTajaU" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agents show ‘psychosis’ in production, but engineering mitigations emerge</strong>: Concerns about ‘AI agent psychosis’ (erratic behavior) are rising, but <a href="http://Monday.com" target="_blank" rel="noopener nofollow">Monday.com</a> has successfully deployed a Slack-integrated AI bot using LLM agents orchestrated via Model Context Protocol (MCP) to provide real-time platform insights, mitigating hallucinations with server-side pre-computation and human-in-the-loop approvals. <a href="https://app.daily.dev/feed-by-ids?id=AJzG7x8FZ&amp;id=AS53pzdrZ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Steam clarifies AI disclosure rules for game developers</strong>: Steam has updated its AI disclosure policy, now requiring developers to only disclose AI-generated content directly consumed by players (e.g., artwork, audio, narrative), exempting AI used for internal workflows or efficiency gains without player-facing impact. <a href="https://app.daily.dev/posts/C2CcXjt7M" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>jQuery 4.0.0 released after a decade</strong>: jQuery 4.0.0, released on its 20th anniversary, focuses on modernization by removing deprecated features, dropping support for older browsers like IE10, and rewriting the codebase as ES modules, resulting in a smaller library size and improved standards compliance. <a href="https://app.daily.dev/feed-by-ids?id=5SOCKxGlX&amp;id=dThF1vPz2" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Wine 11.0 stable release brings major technical upgrades</strong>: Wine 11.0 stable is out with over 6,300 changes, featuring NTSYNC kernel module support for better performance, a completed WoW64 architecture, and upgraded graphics with EGL as the default OpenGL backend and Vulkan 1.4.335 support. <a href="https://app.daily.dev/posts/5LDn1X8et" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>NVIDIA allegedly sought pirated books for AI training</strong>: Apparently, NVIDIA executives, in their infinite wisdom, allegedly contacted Anna’s Archive, a well-known pirate library, to get millions of copyrighted books for AI model training. The lawsuit claims they were explicitly warned about the illegal nature of the content but proceeded anyway, also using other notorious piracy sites like Books3 and Sci-Hub. Because, you know, ethical sourcing is for the little guys, not for companies building the future of AI. <a href="https://app.daily.dev/posts/2IcgLruSE" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI coding tools: the new productivity killer or savior?</strong>: Some developers are touting AI coding tools like Cursor and Claude Code as productivity boons, claiming they can accelerate tedious tasks and scaffold apps rapidly. Yet, others are finding that these tools generate lower-quality code for complex tasks, requiring substantial human correction and leading to ‘burnout’ from managing AI’s failures. It seems the ‘AI coding requires managers’ mantra is less about managing junior devs and more about babysitting a glorified autocomplete that occasionally hallucinates a masterpiece. <a href="https://app.daily.dev/feed-by-ids?id=3DI7vnha4&amp;id=ATZPMmQn8&amp;id=DdiXGTqPN&amp;id=L2hSWhbgY&amp;id=LJqUceZbE&amp;id=Y8gY63ibF&amp;id=aGFVTrinz&amp;id=aKVj0i6Bm&amp;id=j91dGf3TP&amp;id=tJZLXej6S" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>OpenCode and Anthropic’s API drama: who’s really being open?</strong>: OpenCode, the ‘free alternative’ to Claude Code, promotes transparency and community-driven development, letting users link any AI model via API. But then, Oh My OpenCode, a third-party tool, bypassed Anthropic’s API limits by ‘spoofing client headers’ to get free Claude Max access. Anthropic, in a fit of pique, banned accounts without warning. So, ‘open’ until it costs them money, then it’s ‘ban first, ask questions never.’ Classic. <a href="https://app.daily.dev/feed-by-ids?id=15Mjf12HP&amp;id=3RoDG910s" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
