<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 integrate operational database capabilities with their existing data lakehouse architecture, offering a serverless OLTP database that separates compute from storage. Lakebase supports up to 8TB per instance, includes Postgres 17 with pgvector, and is currently GA on AWS, with Azure in public preview and GCP planned for later in 2026. <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</strong>: The Linux kernel 7.0 rc1 has been released, with Linus Torvalds humorously attributing the major version bump to his confusion with large numbers, not a significant feature milestone. This release solidifies Rust support, brings faster cache clearing, non-disruptive kernel updates, and updated AMD/Intel silicon support. It also includes performance improvements for RISC-V and LoongArch, and removes a legacy IBM ThinkPad modem driver. <a href="https://app.daily.dev/posts/cYP8biu57" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google embedding model deprecation causes vendor lock-in</strong>: Google deprecated its <code>text-embedding-004</code> model, rendering thousands of stored vectors unusable for a production RAG-based movie recommender. This highlights a critical issue: different embedding models produce incompatible vector spaces, creating a hidden vendor lock-in. The solution involved re-embedding the entire dataset and using a blue/green swap for the Qdrant collection, emphasizing the need to design for change when persisting embeddings at scale. <a href="https://app.daily.dev/posts/xhDI1zj4i" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Kubernetes 1.35 enhances AI/ML support</strong>: Kubernetes v1.35, codenamed ‘Timbernetes’, introduces features like workload-aware scheduling and stable in-place Pod resource resize, strengthening its role as an operational substrate for AI/ML workloads. <a href="https://app.daily.dev/posts/mAyDHBOpw" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>VS Code gets a fresh file explorer</strong>: A new VS Code extension, ‘Fresh File Explorer’, replaces the standard file explorer to display recently modified files, integrating Git history and pending changes with a heatmap for recency. <a href="https://app.daily.dev/posts/pQedWHBRB" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Fileless privilege escalation in hybrid AD environments</strong>: A red team case study details a 13-step fileless privilege escalation attack chain in hybrid Active Directory environments, allowing a low-privileged user to gain Domain Admin and Azure AD Global Admin access. <a href="https://app.daily.dev/posts/u1TVqzV8C" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>UMMC hit by ransomware attack</strong>: The University of Mississippi Medical Center (UMMC) suffered a ransomware attack, forcing the shutdown of all 35 clinics, compromising their EHR and IT systems, and leading to canceled elective procedures. <a href="https://app.daily.dev/posts/44GuQSR8k" 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 Docker. <a href="https://app.daily.dev/posts/TqgsBjEoV" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Rust RwLock can be slower than Mutex for read-heavy workloads</strong>: An investigation revealed 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>Elastic Cloud supports AWS Graviton4 ARM profiles</strong>: Elastic Cloud Hosted now supports AWS Graviton4-powered ARM hardware profiles, offering up to 40% better price-performance for CPU-bound and data-intensive workloads in select AWS regions. <a href="https://app.daily.dev/posts/0oyHKk4h1" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Anthropic’s Claude compiler video: more marketing than magic</strong>: Anthropic released a marketing video showcasing Claude autonomously building a C compiler capable of compiling Linux and Doom. Critics, however, claim the video used “dishonest framing,” suggesting the actual goal was a stress test of Claude’s extended autonomous operation, not a straightforward demonstration of compiler-building prowess. So, another day, another AI marketing video that stretches the truth further than a rubber band. <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’—pigeons with neural implants, solar panels, and cameras—to guide them along preset routes for infrastructure inspection and search-and-rescue. They claim these birds can cover 400 km daily without battery swaps and are “privacy compliant.” Because nothing says “privacy” like a pigeon with a brain implant and a camera, right? What could possibly go wrong with weaponized wildlife? <a href="https://app.daily.dev/posts/m7v8MA0Dh" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Meta patents AI to simulate deceased users</strong>: Meta was granted a patent for an AI system that simulates a user’s social media activity, including posting and chatting, using an LLM trained on their past data, specifically for when a user is ‘absent’ from the platform. Because what everyone needs is a digital zombie of themselves, posting away after they’ve logged off for good. Ethical concerns? Mental health concerns? Nah, just keep that engagement going! <a href="https://app.daily.dev/posts/AeW0pHCcd" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
