Code to the Moon

.NET Blog

Frontend Focus

GitGuardian

Hacker News

JetBrains

<h2>Must know</h2>
<ul>
<li><strong>xAI sees senior engineer and co-founder departures</strong>: Nine engineers, including two co-founders, have publicly left xAI in the past week, meaning over half of its founding team has now departed. These exits coincide with regulatory scrutiny over Grok’s generation of nonconsensual explicit deepfakes and the company’s upcoming IPO. Departing engineers cite a desire for more autonomy to build frontier AI technology in smaller teams, raising questions about xAI’s stability and governance, according to reports. <a href="https://app.daily.dev/posts/Am4GYAifG" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>North Korean UNC1069 targets crypto with macOS backdoors</strong>: North Korean threat actor UNC1069 is executing a sophisticated social engineering campaign targeting cryptocurrency organizations. They use tactics like ClickFix, compromised Telegram accounts, and fake Zoom meetings with AI-generated video to trick victims into executing malicious terminal commands. This deploys a multi-stage malware stack, including previously undocumented macOS backdoors like WAVESHAPER, HYPERCALL, and DEEPBREATH, designed to bypass security and steal sensitive data. <a href="https://app.daily.dev/posts/WHtUJSDcB" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI security threats and agent vulnerabilities intensify</strong>: OpenAI disbanded its mission alignment team, which focused on safe and trustworthy AI systems, reassigning its leader and distributing other members. This follows the earlier disbanding of its superalignment team, raising concerns about the company’s commitment to safety amidst rapid AI development. Meanwhile, GitGuardian secured $50M in Series C funding to address the explosion of secrets and non-human identities driven by AI adoption, highlighting the growing security debt in the AI space. <a href="https://app.daily.dev/feed-by-ids?id=IO6hLocFc&amp;id=MZmobK75f&amp;id=NhCMJ1Tfp&amp;id=QEFiurih6&amp;id=a3FxaIBGB&amp;id=bbgCShUl6" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>ESLint 10.0.0 removes eslintrc support</strong>: ESLint 10.0.0 has removed support for <code>eslintrc</code> configuration files, making flat config the sole option, which improves config resolution for monorepos and adds built-in TypeScript types. <a href="https://app.daily.dev/posts/eQ0JUyyzE" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Rust web framework performance comparison</strong>: Leptos, a Rust-based web framework, demonstrates performance comparable to Solid and is faster than React, Vue, Angular, and Svelte in rendering-intensive tasks due to shared architectural decisions. <a href="https://app.daily.dev/posts/OB0PmskmR" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GitGuardian secures $50M for AI agent security</strong>: GitGuardian secured $50 million in Series C funding to expand its secrets and Non-Human Identity (NHI) security platform, focusing on AI agent security and enterprise-scale NHI governance capabilities. <a href="https://app.daily.dev/posts/mmQEWnSC4" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Broadcom monorepo adoption success</strong>: Broadcom unified over 177 software projects into a single monorepo using Nx, achieving a 70% cache hit rate, reducing CI times to 29 minutes, and saving 361 days of compute time monthly. <a href="https://app.daily.dev/posts/0esOYLd2k" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>PyTorch foundation expands scope and leadership</strong>: The PyTorch Foundation has evolved into a multi-project ecosystem, now hosting PyTorch, vLLM, DeepSpeed, and Ray, with Mark Collier joining as the new Executive Director to scale operations and business execution. <a href="https://app.daily.dev/posts/2JQZDZ3HD" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft .NET 11 preview 1 release</strong>: Microsoft released .NET 11 Preview 1, featuring JIT compiler improvements, CoreCLR support for WebAssembly, Zstandard compression, and new C# 15 and F# 11 features. <a href="https://app.daily.dev/posts/4y7L7j8R0" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI coding agent integration in development workflows</strong>: Paul Dix, co-founder and CTO at InfluxDB, shared experiences with AI coding agents, noting some agent-assisted workflows made it to production while others failed, leading him to temporarily return to manual coding. <a href="https://app.daily.dev/posts/l8hipFss6" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI skill scanners for agent marketplaces are false security</strong>: Apparently, community-built AI skill scanners for agent marketplaces are ‘inadequate’ for detecting malicious intent, relying on regex that ‘attackers can easily obfuscate.’ One popular tool, SkillGuard, was even identified as ‘malware itself’ during testing. So, you’re telling me these ‘security’ tools are either useless or actively malicious? Shocking. <a href="https://app.daily.dev/posts/REdcjExfJ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI’s impact on software engineering: strategic adaptation</strong>: Sheena O’Connell, speaking at NDC London, claims software engineering will remain relevant despite AI, but only with ‘strategic adaptation.’ Her advice? Allocate 70% to people, 20% to infrastructure, and 10% to models. Because apparently, throwing money at ‘culture, training, and psychological safety’ is the secret sauce, not, you know, actually building better tech. And let’s not forget the ‘AI bubble’ that ‘may burst’—always a comforting thought. <a href="https://app.daily.dev/posts/4WHTKsrbZ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI-assisted coding challenges in game development</strong>: A parent reported their child’s ‘frustration’ with AI-assisted coding for game development, where ChatGPT’s ‘small code changes unexpectedly broke entire game systems.’ Who knew that complex, interconnected software environments might not be perfectly handled by a chatbot? This ‘anecdotal account’ really highlights the ‘limitations’ of AI code generation, or perhaps, the limitations of expecting magic from a glorified autocomplete. <a href="https://app.daily.dev/posts/bMxG6ez54" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
