<h2>Must know</h2>
<ul>
<li><strong>Google releases powerful new gemini model</strong>: Google just dropped Gemini 3.1 Pro in public preview, touting it as their most powerful agentic and coding model yet. It scored 77.1% on the ARC-AGI-2 benchmark, a huge jump from its predecessor, and features a massive 1 million token context window. This model is available across various platforms, including GitHub Copilot and Google AI Studio, and is priced competitively against rivals like Anthropic’s Opus 4.6, according to Google’s announcement. <a href="https://app.daily.dev/posts/PjE6Xo1XG" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>MySQL governance under oracle challenged by community</strong>: Over 248 database professionals, including folks from Percona and MariaDB, signed an open letter urging Oracle to hand over MySQL governance to an independent non-profit. They’re concerned about declining commit activity, a shrinking contributor base, and MySQL’s lack of AI-era features. The community points to PostgreSQL’s success with community-led governance as a model, but analysts are skeptical Oracle will give up control easily, especially with trademarks involved. <a href="https://app.daily.dev/posts/GY7vcPtOu" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agents are reshaping software development</strong>: Agentic AI is fundamentally changing how software gets built, moving beyond simple prompt-response to autonomous agents that plan and execute. Tools like Kiro and Claude Code are boosting engineer productivity, making non-adoption a career risk. McKinsey data shows 23% of organizations are already scaling agentic AI, with new roles like ‘Agent Orchestration Engineer’ emerging, but concerns around bias, compliance, and transparency remain, as highlighted by various industry reports. <a href="https://app.daily.dev/feed-by-ids?id=037a5YIjx&amp;id=4CjaWv3pP&amp;id=7XxAvY7Df&amp;id=Bl6akHr3q&amp;id=DqRWRtC6t&amp;id=F8eGl7g0G&amp;id=LCg5ueVO7&amp;id=SBASpAo7A&amp;id=UJqBwwcg1&amp;id=Uw1m3wdti" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>AI memory and agent management solutions emerge</strong>: New solutions like Mengram’s open-source AI memory API and Reload’s Epic platform are tackling AI agent memory and collaboration, with Mengram focusing on semantic, episodic, and procedural memory, and Reload providing shared long-term memory for AI coding teams. <a href="https://app.daily.dev/feed-by-ids?id=0tYkAozTe&amp;id=MQxvV1b9u" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI infrastructure development and energy demands</strong>: Cisco and NVIDIA are partnering on the Secure AI Factory, a modular architecture for AI compute, networking, storage, and security, while SoftBank plans a massive $33 billion natural gas power plant to meet the surging energy demands of AI data centers. <a href="https://app.daily.dev/feed-by-ids?id=IYsqAR4Cy&amp;id=VBQEBjah6" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI and robotics industrial adoption accelerates</strong>: Toyota is deploying Agility Robotics’ Digit humanoid robots in its RAV4 plant for auto parts unloading, and NVIDIA’s survey shows 90% of telecom respondents report AI increasing revenue and reducing costs, indicating a strong industrial push for AI and robotics. <a href="https://app.daily.dev/feed-by-ids?id=WnnenuOfb&amp;id=p1OVwpWSR" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Redwood materials expands energy storage for AI data centers</strong>: Redwood Materials, backed by Google and Nvidia, is rapidly expanding its energy storage business unit, repurposing EV batteries to meet the surging power demands of AI data centers, which face 5+ year grid connection wait times. <a href="https://app.daily.dev/posts/1mD3jO5pQ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Hugging face launches community evals for AI model benchmarking</strong>: Hugging Face introduced “Community Evals,” a beta feature aiming to standardize and democratize AI model benchmarking through community-driven leaderboards and reproducible evaluation specifications. <a href="https://app.daily.dev/posts/9WqEmBh9G" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Apache Polaris graduates to top-level project</strong>: Apache Polaris, an open-source catalog for Iceberg and data lakehouse interoperability, graduated to a Top-Level Project under the Apache Software Foundation, signaling mature governance and momentum for open data lakehouse standards. <a href="https://app.daily.dev/posts/ovRl52TUb" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Vitest vs jest 2026 performance benchmarks</strong>: Vitest 3.x significantly outperforms Jest 30 in performance benchmarks from a 50,000-test monorepo, showing 5.6x faster cold starts and 28x faster watch mode re-runs due to its Vite-based transform pipeline. <a href="https://app.daily.dev/posts/T3G4VUzdk" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Anthropic’s claude pro integrates with microsoft powerpoint</strong>: Oh, joy! Claude Pro now integrates with Microsoft PowerPoint. Because what we really needed was an AI to help us make even more bullet-point-laden presentations. The promotional announcements claim Claude Pro is “getting bigger every week,” which I’m sure is a totally unbiased assessment. Can’t wait for AI to generate my next deck, complete with perfectly bland corporate jargon. <a href="https://app.daily.dev/posts/oLQ4V0bPw" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI code quality and production visibility concerns</strong>: Apparently, AI coding tools are shipping JavaScript so fast that developers can’t review it, leading to “subtle production bugs.” The solution? “Production visibility” tools like TrackJS, which, coincidentally, is the product the article is shilling. So, AI makes a mess, and then we need <em>more</em> tools to clean up the AI’s mess. Sounds like a sustainable business model for everyone but the actual engineers. <a href="https://app.daily.dev/feed-by-ids?id=2Xamq3Lpe&amp;id=IR0A3b0XK" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI’s impact on productivity and employment: european and indian perspectives</strong>: A study says AI boosts European labor productivity by 4% with no job losses, while an Indian think tank says AI isn’t an immediate threat to IT jobs. Yet, some Indian firms expect headcount to shrink. And then there’s the opinion piece calling AI job-loss fears “historical mistakes.” So, AI is simultaneously a job creator, a job reducer, and a non-factor, depending on who you ask. Clear as mud, as always. <a href="https://app.daily.dev/feed-by-ids?id=iZlTkI8Me&amp;id=tiBubXcsb&amp;id=yOsJJlumD" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
