<h2>Must know</h2>
<ul>
<li><strong>Rust’s true value is correctness, not just memory safety</strong>: Rust’s primary strength lies in its type system, which enforces correctness through features like enums, pattern matching, and the <code>Result</code> type, catching errors at compile time. This approach, highlighted by Helge Penne at NDC TechTown, reduces the need for extensive unit testing and makes refactoring safer, ensuring that code that compiles is more likely to work as intended. It’s a fundamental shift from traditional languages where memory safety is often the sole focus. <a href="https://app.daily.dev/posts/Xs6GvdOdx" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Bun runtime gets faster and more feature-rich</strong>: Bun v1.3.6 introduces new APIs for tarball creation/extraction and JSONC parsing, alongside significant performance boosts. <code>Response.json()</code> is now 3.5x faster, <code>async/await</code> 15% faster, and <code>Promise.race</code> 30% faster. This release also brings SIMD-optimized <code>Buffer.indexOf</code>, HTTP/HTTPS proxy support, and 45 bug fixes, further solidifying Bun’s position as a high-performance JavaScript runtime, as reported by the official Bun blog. <a href="https://app.daily.dev/posts/Wx4br4Zyw" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google Gemini calendar prompt injection exposed enterprise data</strong>: A security flaw in Google Gemini allowed attackers to extract private calendar data by embedding malicious natural language prompts in Google Calendar invitations. When users queried Gemini about their schedules, the AI processed these prompts, bypassing privacy controls. Google has since patched the vulnerability, but the incident highlights the emerging security risks of indirect prompt injection in LLM-powered applications, especially in enterprise environments, as detailed by security researchers. <a href="https://app.daily.dev/posts/dHjNBMkFO" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Rust-analyzer 0.3.2761 improves developer experience</strong>: The latest Rust-analyzer update, version 0.3.2761, brings 13 fixes and improvements, including better flycheck triggering for non-workspace files, enhanced symbol indexing for macros, and improved completion suggestions for <code>mut</code> and <code>raw</code> keywords. <a href="https://app.daily.dev/posts/kEIknGK9T" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Go agent ports Rust template engine with AI</strong>: An experienced developer successfully ported MiniJinja, a Rust-based Jinja2 template engine, to native Go using AI agents, requiring only 45 minutes of active guidance and 10 hours of autonomous agent work, leveraging snapshot tests for validation. <a href="https://app.daily.dev/posts/VOMBPE1Xs" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Next.js 16 boosts performance with Turbopack and AI debugging</strong>: Next.js 16 stabilizes Turbopack as the default bundler for 2-5x faster production builds, introduces explicit opt-in caching with ‘use cache’, and integrates AI-assisted debugging via Model Context Protocol (MCP) in DevTools. <a href="https://app.daily.dev/posts/91xiEmcVe" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Cloudflare acquires Astro web framework, keeps it open source</strong>: Cloudflare has acquired The Astro Technology Company, but the Astro web framework will remain open source, MIT-licensed, and platform-agnostic, with the entire team joining Cloudflare to continue development. <a href="https://app.daily.dev/feed-by-ids?id=a23rECuU6&amp;id=6Yvvs8a9F&amp;id=yRAOJZmiR" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Electron 40.0.0 released with Chromium 144 and Node 24.11.1</strong>: Electron 40.0.0 is out, featuring upgrades to Chromium 144.0.7559.60, V8 14.4, and Node 24.11.1, bringing HDR color space support, a hardware acceleration detection API, and dynamic ESM imports in preloads. <a href="https://app.daily.dev/posts/VsVvB0iFk" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft open sources XAML Studio for .NET UI prototyping</strong>: Microsoft has open-sourced XAML Studio under the .NET Foundation, aiming to accelerate prototyping and feedback loops for XAML-based UI development, with the latest version (2.0) featuring a Fluent UI redesign and enhanced WinUI 3 integration. <a href="https://app.daily.dev/posts/hUJrXK08w" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AWS European Sovereign Cloud now available for EU data residency</strong>: AWS European Sovereign Cloud is now generally available, with its first region in Brandenburg, Germany, offering physically and logically separate infrastructure operated solely by EU residents to meet strict data residency and operational autonomy requirements. <a href="https://app.daily.dev/feed-by-ids?id=DTVSjPW7g&amp;id=B8P78hRXl" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI coding tools: the productivity paradox continues</strong>: Despite the hype, AI coding assistants are still generating 46% more code but doubling bugs and increasing code churn. Developers using AI heavily are reportedly 19% slower, and PR review times are up by 91%. It seems AI optimizes for typing speed, not actual engineering, leading to more technical debt and developers merging code they don’t even understand. Who needs architectural integrity when you have more lines of code, right? <a href="https://app.daily.dev/posts/5khP9Wt37" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agents: the new frontier of security vulnerabilities</strong>: Anthropic’s Cowork, designed for ‘non-technical users,’ has an unpatched prompt injection vulnerability that allows file exfiltration without user approval. Apparently, Anthropic’s solution is to tell users to ‘monitor suspicious activity’—because non-technical users are famously good at spotting zero-day exploits. This is what happens when you rush AI to market without basic security hygiene, creating a new class of ‘AI-enhanced’ insider threats. What could possibly go wrong? <a href="https://app.daily.dev/feed-by-ids?id=s11rN52dz&amp;id=jaKSkIh92&amp;id=rAHVvRPEj" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Small language models: the new ‘disruptor’ of the week</strong>: Enterprises are suddenly ‘favoring’ Small Language Models (SLMs) because LLMs are ‘liabilities’ due to cost and latency. Apparently, 7-billion parameter models can now ‘outperform’ frontier LLMs on reasoning, while being 10,000 times smaller. This narrative, pushed by unnamed research and industry commentators, conveniently ignores the fact that these SLMs are often used in ‘cascading’ architectures that still rely on those ‘liability’ LLMs for complex tasks. It’s almost like the industry just found a new buzzword to justify cheaper compute while still riding the AI hype train. <a href="https://app.daily.dev/feed-by-ids?id=HX43N3MkE&amp;id=J1BKrq2u6" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
