<h2>Must know</h2>
<ul>
<li><strong>JetBrains IDEs get native OpenAI Codex integration</strong>: OpenAI Codex is now natively integrated into JetBrains IDEs, including IntelliJ, PyCharm, WebStorm, and Rider, starting with version 2025.3. Powered by GPT-5.2-Codex, this AI agent supports refactoring, code migrations, and security vulnerability detection with adjustable autonomy levels and ‘reasoning budgets.’ Users can access it via a JetBrains AI subscription, ChatGPT account, or OpenAI API key, with a limited-time free access period, as reported by JetBrains. <a href="https://app.daily.dev/posts/uAvdGTErT" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GitHub Copilot expands with multi-platform SDK</strong>: GitHub has released a multi-platform SDK for integrating Copilot’s agentic workflows into applications, supporting Python, TypeScript, Go, and .NET. This SDK enables embedding the same engine used by the Copilot CLI, offering functionalities like orchestration, planning, and programmatic file edits. It’s currently in technical preview and requires a GitHub Copilot subscription, signaling GitHub’s push to expand Copilot’s ecosystem. <a href="https://app.daily.dev/feed-by-ids?id=GXXnGjqMp&amp;id=GuG7K2ryZ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agents pose risks in IT operations automation</strong>: AI agents introduce non-deterministic actions into IT operations, increasing risk, similar to ‘cowboy’ sysadmins making unreproducible changes. While useful for exploratory tasks like generating playbooks and Dockerfiles, direct modification of live production systems by AI is deemed unsafe. The recommendation is for AI to propose changes to version-controlled artifacts, emphasizing human review, testing, and reproducibility before deployment, as highlighted in a recent analysis. <a href="https://app.daily.dev/posts/U8PYVjc73" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Rust 1.93.0 updates musl libc and inline assembly</strong>: Rust 1.93.0, released on January 22, 2026, updates the bundled musl libc to version 1.2.5, improving DNS resolution for static Linux binaries, and enhances inline assembly handling for more efficient code integration. <a href="https://app.daily.dev/feed-by-ids?id=wF6DW7Hp4&amp;id=O9g10PE7Q" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Vercel Sandboxes introduces filesystem snapshots</strong>: Vercel Sandboxes now supports filesystem snapshots, allowing users to capture and launch new sandbox environments from saved states, streamlining workflows for parallel testing and rapid prototyping. <a href="https://app.daily.dev/posts/Uv3Zx9otG" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Feast integrates with PyTorch for feature management</strong>: Feast, an open-source feature store, has joined the PyTorch Ecosystem to address training-serving skew in production AI systems, providing a unified API for consistent feature management across model training and deployment. <a href="https://app.daily.dev/posts/dPG1hvjH4" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Datadog launches automatic retention filters for RUM and APM</strong>: Datadog released cross-product retention filters that automatically retain Application Performance Monitoring (APM) traces linked to Real User Monitoring (RUM) sessions, unifying frontend and backend observability data for improved troubleshooting. <a href="https://app.daily.dev/posts/fs5040Sq6" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Pulumi adds replacementTrigger for resource recreation</strong>: Pulumi v3.215.0 introduces the <code>replacementTrigger</code> option, enabling automatic resource recreation when specified metadata values change, useful for tasks like cycling secrets or updating resources based on external version numbers. <a href="https://app.daily.dev/posts/ay0AhT4GQ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GitHub Actions offers 1 vCPU Linux runners</strong>: GitHub Actions now provides 1 vCPU Linux runners (ubuntu-slim) as a generally available, lower-cost option for lightweight automation tasks, offering 1 vCPU and 5GB RAM with a 15-minute execution limit. <a href="https://app.daily.dev/posts/2Mmk9fDWb" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AMD prepares Linux patches for next-gen EPYC features</strong>: AMD has submitted 19 Linux kernel patches to support new CPU features like Global Bandwidth Enforcement (GLBE), Global Slow Bandwidth Enforcement (GLSBE), and Privilege Level Zero Association (PLZA) for upcoming EPYC ‘Venice’ processors, enhancing Quality of Service. <a href="https://app.daily.dev/posts/07cLLCzjH" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI agents are here to write your code, just don’t ask them to be smart</strong>: Apparently, ‘Claude Code with Opus 4.5 has evolved to handle most development work without additional frameworks or tooling,’ according to some YouTube guru. It’s got ‘plan mode, spec-driven development, sub-agents, and skills’—all the buzzwords. But then, in the same breath, they admit ‘developer judgment, product thinking, and strategic decision-making remain critical skills that AI cannot replace.’ So, it’s all you need, except for the actual thinking part. Got it. <a href="https://app.daily.dev/posts/oecQfT5G1" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI will make your JavaScript ‘10x faster’ and your brain ‘10x dumber’</strong>: Some folks are claiming LLMs can optimize JavaScript performance ‘from hours or days to minutes,’ even turning a game from 20-30 FPS to 60 FPS. The catch? ‘Some knowledge transfer is sacrificed.’ So, you get faster code, but you lose the understanding of <em>why</em> it’s faster. Sounds like a great way to build a codebase no human can debug, all while patting ourselves on the back for ‘shipping projects without accumulating technical debt.’ <a href="https://app.daily.dev/posts/NttZ4o4xB" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agents are the new ‘cowboy sysadmins,’ but with more buzzwords</strong>: We’re being told that AI agents are going to revolutionize IT operations, but also that they introduce ‘non-deterministic actions’ and are ‘unsafe’ for direct production modifications. The solution? Have AI propose changes to version-controlled systems, then have humans review, test, and reproduce them. So, we’re just adding an extra, potentially unpredictable, step to the existing process, and calling it ‘innovation.’ What could possibly go wrong? <a href="https://app.daily.dev/posts/U8PYVjc73" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
