<h2>Must know</h2>
<ul>
<li><strong>Databricks launches Lakebase for AI workloads</strong>: Databricks has launched Lakebase, a new PostgreSQL-compatible database specifically designed for AI workloads. It aims to merge operational database capabilities with their existing data lakehouse architecture, offering a serverless OLTP solution that separates compute from storage. Currently available on AWS, with Azure in public preview and GCP planned, Lakebase supports features like instant data branching and pgvector, making it relevant for real-time ML feature serving and AI agent memory. <a href="https://app.daily.dev/posts/56RHAShlB" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Linux kernel 7.0 rc1 released with Rust support</strong>: The Linux kernel 7.0 rc1 has been released, with Linus Torvalds humorously attributing the major version bump to his dislike of large numbers, rather than a significant feature overhaul. This release solidifies Rust support, improves cache clearing, and introduces non-disruptive kernel updates. It also includes updated AMD/Intel silicon support and performance enhancements for RISC-V and LoongArch, as reported by multiple sources. <a href="https://app.daily.dev/feed-by-ids?id=cYP8biu57&amp;id=8ssuMcdSU&amp;id=JXfUDc5F4&amp;id=QdVE1yqfx&amp;id=TcabAvodN&amp;id=cQun6gD6y&amp;id=zaDpYuaxU" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>OpenAI’s Codex team uses AI for everything</strong>: OpenAI’s Codex team operates with minimal hierarchy and deeply integrates AI into every workflow, from onboarding to PR reviews and large-scale refactors. They use AI for planning and user feedback triage, emphasizing small, autonomous squads and rapid iteration. This approach highlights a practical, AI-first development methodology within a leading AI research organization, as detailed in a conversation with Engineering Lead Thibault Sottiaux. <a href="https://app.daily.dev/posts/8TOu6Vf1U" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Fresh File Explorer for VS Code</strong>: A new VS Code extension, Fresh File Explorer, replaces the standard file explorer to display recently modified files, integrating Git history and offering features like a recency heatmap and one-click restore for deleted files. <a href="https://app.daily.dev/posts/pQedWHBRB" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Kubernetes 1.35 strengthens AI/ML support</strong>: Kubernetes v1.35, codenamed ‘Timbernetes,’ introduces features like workload-aware scheduling and stable in-place Pod resource resize, further solidifying its role as an operating system for AI/ML workloads. <a href="https://app.daily.dev/posts/mAyDHBOpw" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Shuru: microVM sandbox for AI agents on macOS</strong>: Shuru is a new local-first microVM sandbox for macOS, designed for AI agent workloads, providing secure, ephemeral, and reproducible environments using Apple’s Virtualization.framework without requiring Docker. <a href="https://app.daily.dev/posts/TqgsBjEoV" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Non-coding tasks kill developer productivity</strong>: A developer’s personal account highlights that a 10-second code fix can take over 10 hours due to non-coding tasks like fighting sandboxed environments, CUDA conflicts, and build system issues, which current LLM tools don’t address. <a href="https://app.daily.dev/posts/OBIqfWI7x" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Rust RwLock underperforms Mutex in read-heavy loads</strong>: A hands-on investigation found that <code>RwLock</code> performed significantly worse than <code>Mutex</code> for a read-heavy LRU tensor cache in Rust due to cache-line ping-pong, suggesting <code>Mutex</code> might be better for very short critical sections. <a href="https://app.daily.dev/posts/rFMs5918e" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft Project Silica advances glass storage</strong>: Microsoft’s Project Silica has advanced glass-based data storage, extending it from expensive fused silica to common borosilicate glass, with accelerated aging tests suggesting data longevity of at least 10,000 years. <a href="https://app.daily.dev/posts/KqLBpGspa" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>FreeBSD Rust kernel support nearing stability</strong>: FreeBSD’s Q4 2025 status report indicates that Rust kernel programming interfaces (KPIs) are expected to be stable enough for developers to experiment with writing new kernel code by early 2026. <a href="https://app.daily.dev/posts/H5Wur3zjw" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Anthropic’s Claude compiler video: more hype than reality?</strong>: Anthropic released a marketing video showcasing Claude autonomously building a C compiler capable of compiling Linux and Doom. Critics, however, are calling out the “dishonest framing,” suggesting the real goal was merely a stress test of Claude’s ability to run autonomously for an extended period, not a straightforward demonstration of compiler-building prowess. Because, apparently, we’re all supposed to believe AI can just whip up a compiler for Doom on a whim. <a href="https://app.daily.dev/posts/vGYk0DbTO" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Neiry’s ‘bio-drones’: pigeons with neural implants</strong>: Neiry, a deeptech company, is testing ‘bio-drones’—live pigeons with neural implants, solar panels, and cameras—to guide them along preset routes for infrastructure inspection and search-and-rescue. They claim “privacy compliance” through on-device processing, which sounds suspiciously like a fancy way of saying “we’ll filter out the embarrassing bits.” Because nothing says cutting-edge tech like strapping a camera to a bird and calling it a drone. <a href="https://app.daily.dev/posts/m7v8MA0Dh" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Real-time glint particle rendering: more sparkles for everyone</strong>: A new research technique for real-time rendering of microscopic glint particles claims to render millions of sparkles on surfaces at over 280 FPS on consumer NVIDIA GPUs. It generates particle positions on-the-fly, eliminating the need to store every particle. Because what the world really needs is more realistic digital glitter, and apparently, 337 lines of code is all it takes to achieve this groundbreaking feat. <a href="https://app.daily.dev/posts/rT496vKCA" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
