<h2>Must know</h2>
<ul>
<li><strong>Google releases Gemini 3.1 Pro, Anthropic updates Claude policy</strong>: Google has launched Gemini 3.1 Pro in public preview, touting it as their most powerful agentic and coding model with a 1 million token context window and multimodal capabilities. It scored 77.1% on the ARC-AGI-2 benchmark, significantly outperforming previous versions and competitors like Anthropic’s Opus 4.6. Meanwhile, Anthropic updated its Claude Code documentation, restricting OAuth token usage from Pro/Max subscription accounts in third-party tools, causing community backlash and raising concerns about the future of affordable flat-rate AI subscriptions for commercial development. This move suggests a push towards API key authentication for developers building products, impacting tools like OpenClaw and NanoClaw. <a href="https://app.daily.dev/feed-by-ids?id=PjE6Xo1XG&amp;id=3lsZ5Vula&amp;id=FQxtQgg9R&amp;id=GYB8zPRsi" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>MySQL governance under Oracle challenged by community</strong>: Over 248 database professionals, including engineers from Percona and MariaDB, have signed an open letter urging Oracle to transfer MySQL governance to an independent non-profit foundation. Concerns include declining commit activity, a shrinking contributor pool, lack of AI-era features, and an opaque roadmap. This community push highlights a desire for a more open, PostgreSQL-like development model, though analysts are skeptical about resolving power dynamics if Oracle retains trademark control. <a href="https://app.daily.dev/posts/GY7vcPtOu" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI infrastructure demands massive energy and investment</strong>: The rapid expansion of AI is driving unprecedented demand for infrastructure, both digital and physical. Cisco and NVIDIA are partnering on the Secure AI Factory, a modular architecture for AI compute, networking, storage, and security, anticipating a $267 billion partner-delivered AI services market by 2030. Concurrently, SoftBank’s SB Energy plans a $33 billion, 9.2 gigawatt natural gas power plant in the US, potentially to fuel AI data centers, highlighting the immense energy requirements and environmental concerns associated with scaling AI. <a href="https://app.daily.dev/feed-by-ids?id=IYsqAR4Cy&amp;id=VBQEBjah6&amp;id=1mD3jO5pQ" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>AI agents reshape software development workflows</strong>: AI tools like Kiro and Claude Code are shifting software development from prompt-response to autonomous agents that plan, execute, and adapt, with McKinsey data indicating 23% of organizations are scaling agentic AI. <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>
<li><strong>Kafka 4.2.0 introduces share groups and stream improvements</strong>: Apache Kafka 4.2.0 is out with production-ready Share Groups for extended processing and adaptive batching, alongside GA status for Kafka Streams’ server-side rebalance protocol and dead letter queue support. <a href="https://app.daily.dev/posts/0DcTaIDm9" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Vitest 3.x significantly outperforms Jest 30 in benchmarks</strong>: Vitest 3.x shows dramatic performance gains over Jest 30, with 5.6x faster cold starts and 28x faster watch mode re-runs, making it a strong contender for projects already using Vite. <a href="https://app.daily.dev/posts/T3G4VUzdk" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI memory and agent management solutions emerge</strong>: Mengram launched an open-source AI memory API with semantic, episodic, and procedural memory types, while Reload’s Epic provides shared long-term memory for AI coding agents, addressing current AI agent limitations. <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>Tailscale workload identity federation is generally available</strong>: Tailscale’s workload identity federation is now GA, allowing CI/CD pipelines and cloud workloads to authenticate with short-lived OIDC tokens, eliminating static credentials and integrating with Terraform. <a href="https://app.daily.dev/posts/ekx9IIs1v" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Spring Boot 4.0.3 released</strong>: Spring Boot 4.0.3, a maintenance release in the 4.0.x line, is now available from the Spring project at Broadcom. <a href="https://app.daily.dev/posts/4wdP32pyI" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Node.js memory usage halved with node-caged</strong>: Cloudflare, Igalia, and the Node project collaborated on <code>node-caged</code>, a Docker image with V8 pointer compression that can halve Node.js memory usage in production. <a href="https://app.daily.dev/posts/Mm7b9KCfF" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI coding tools: the curse of individual productivity</strong>: Wix Engineering found that while Claude Code boosted individual developer output, tools like Cursor showed no measurable gain despite users ‘feeling’ more productive. This ‘AI Curse’ means individual productivity explodes, but organizational coordination and review processes become new bottlenecks. So, you’re faster, but the team’s still stuck. Progress! <a href="https://app.daily.dev/posts/qQhctEUPB" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI slop is the new technical debt</strong>: Open source maintainers are now drowning in ‘AI slop’—bug fixes generated by ‘script kiddies’ who don’t understand the code, creating an extra review burden. Apparently, AI makes it easier to contribute bad code than good. Who knew? <a href="https://app.daily.dev/feed-by-ids?id=dz3RmQVnk&amp;id=j5Q6D03Mc&amp;id=opnnf1fiE&amp;id=uNeZBI42S" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agents: the new buzzword bingo for consultants</strong>: Consulting firms are now claiming AI is ‘rewriting the rules’ by compressing project timelines and shifting value from analysis to outcomes. They’re also evolving their staffing models from pyramids to diamonds. Because, you know, AI is so revolutionary it’s finally making consultants deliver actual outcomes instead of just talking about them. Shocking. <a href="https://app.daily.dev/posts/hriqsZKLs" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
