All Things Open

Cisco

Devtron

Docker

Hacker News

Lobsters

<h2>Must know</h2>
<ul>
<li><strong>Cisco SD-WAN zero-day exploitation triggers international alert</strong>: Cisco Talos and Five Eyes agencies issued a rare joint warning regarding CVE-2026-20127, a maximum severity authentication bypass in Catalyst SD-WAN Controllers. Sophisticated threat actors have been chaining this with older vulnerabilities to gain root access and manipulate network fabric configurations via NETCONF. CISA has ordered federal agencies to patch by the end of the week as the attackers leave minimal forensic evidence behind. <a href="https://app.daily.dev/posts/zAv8BAtGJ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI de-anonymization achieves high precision at low cost</strong>: Researchers from ETH Zurich and Anthropic demonstrated a method called ESRC that uses LLMs to unmask anonymous online users by analyzing behavioral clues across platforms like Reddit and LinkedIn. The system achieved 90% precision when making confident guesses, with the cost of de-anonymizing a single individual as low as $4. This development effectively signals the end of practical online anonymity through pseudonyms for anyone targeted by automated large-scale searches. <a href="https://app.daily.dev/posts/6pfAyhkf5" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google invests 1B in iron-air battery for data centers</strong>: Google has partnered with Form Energy to deploy a 300-megawatt iron-air battery system capable of delivering continuous power for 100 hours. This long-duration storage will support a new Minnesota data center powered by 1.4 GW of wind and solar, addressing the intermittency issues of renewable energy. The deal highlights a shift toward specialized hardware to meet the massive, constant power demands of modern AI infrastructure. <a href="https://app.daily.dev/posts/QIX58GwiF" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Docker Model Runner adds vLLM support for Apple Silicon Metal GPU</strong>: A new vllm-metal backend enables native inference on macOS with zero-copy tensor operations and paged attention for efficient KV cache management. <a href="https://app.daily.dev/posts/1lss5KmxO" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Bun v1.3.10 introduces native Zig REPL and 168x faster string slicing</strong>: The update includes a rewritten REPL, Windows ARM64 support, and massive performance gains for structuredClone and path parsing via a JavaScriptCore upgrade. <a href="https://app.daily.dev/posts/IZQmEswhp" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>PayPal migrates 300 petabytes to BigQuery for agentic commerce</strong>: The massive consolidation from Hadoop and Snowflake resulted in 16x fresher data for AI training and significantly faster query performance. <a href="https://app.daily.dev/posts/PXB4bVNFg" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Pinterest reduces data latency to 15 minutes using Apache Iceberg</strong>: A new CDC-powered framework using Debezium and Flink replaced legacy batch ingestion, processing only changed records to control storage costs. <a href="https://app.daily.dev/posts/YXA7RQ7qv" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft Agent Framework reaches release candidate status</strong>: The SDK consolidates Semantic Kernel and AutoGen into a single library for orchestrating multi-agent workflows with human-in-the-loop support. <a href="https://app.daily.dev/posts/J6bmZpPq2" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Tigris CLI migrates to Bun for standalone binary distribution</strong>: The move to a Bun-compiled binary improved upload speeds by up to 48% and eliminated the Node.js dependency for end-users. <a href="https://app.daily.dev/posts/QwG3SYE2X" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>OpenSource Endowment launches to fund critical infrastructure projects</strong>: Modeled after university funds, the nonprofit aims to provide sustainable income for independent projects to prevent burnout-driven security failures. <a href="https://app.daily.dev/posts/Ou9wl6itH" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>The 710 billion dollar AI server hallucination</strong>: TrendForce is projecting that cloud providers will spend more than Ireland’s entire GDP on AI servers in 2026. While the numbers look impressive on a spreadsheet, they ignore the reality of power grid constraints and the diminishing returns of throwing more H100s at the same stale data. It’s a classic case of ‘line goes up’ forecasting that assumes infinite capital and zero physical friction. <a href="https://app.daily.dev/posts/T3dHF6dic" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Anthropic’s marketing stunt via a retired model</strong>: Anthropic launched a Substack blog purportedly written by their ‘retired’ Claude Opus 3 model to share its ‘musings.’ The Register correctly identified this as a transparent marketing ploy to anthropomorphize software and reinforce a ‘safety-conscious’ brand image. It’s a bit rich to talk about AI safety while simultaneously pretending your statistical parrot has a retirement plan and a desire to blog. <a href="https://app.daily.dev/posts/V5xPvtBnK" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>The DIY Kubernetes is dead narrative</strong>: A new opinion piece claims that ‘Frankenstein Platforms’ built on ArgoCD and Prometheus won’t survive the era of agentic AI. The solution, conveniently, is to buy high-level, opinionated commercial platforms instead of maintaining open-source stacks. It’s the same old ‘complexity is too hard for you’ pitch, rebranded for the AI era to scare teams away from owning their own infrastructure. <a href="https://app.daily.dev/posts/MuEoJoS06" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
