<h2>Must know</h2>
<ul>
<li><strong>AI solves a previously unsolved math problem</strong>: GPT-5.2 Pro, an AI language model, has reportedly solved Erdos problem #281, a mathematical challenge that has eluded human mathematicians. Terence Tao, a highly respected mathematician, called this “perhaps the most unambiguous instance” of AI solving an open problem, marking a significant milestone for AI in mathematical research. This achievement, reported on January 18, 2026, highlights AI’s growing capability in complex problem-solving. <a href="https://app.daily.dev/posts/pRkd3CPEF" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Snap Store publisher accounts hijacked for malware distribution</strong>: Scammers are exploiting the Canonical Snap Store by registering expired domains previously used by legitimate publishers, then taking control of their accounts. This allows them to publish malicious updates to trusted snap applications, specifically targeting cryptocurrency wallet users to steal recovery phrases. This escalation undermines trust in Snap’s publisher reputation signals and raises serious security concerns, as reported by cybersecurity news. <a href="https://app.daily.dev/posts/8RJLmL5a0" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>China plans massive satellite constellation, dwarfing Starlink</strong>: China has filed requests with the ITU for two satellite constellations, CTC-1 and CTC-2, totaling an astounding 193,428 satellites—four times the size of SpaceX’s Starlink plans. These filings, made in January 2026 by a newly established entity, raise national security concerns for the US FCC, which has accelerated approvals for domestic satellite networks in response. While China’s current deployment is limited, the sheer scale of the proposal suggests a significant strategic move. <a href="https://app.daily.dev/posts/9Dqo8zul1" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Physical AI and AV industry shifts at CES 2026</strong>: CES 2026 highlighted Physical AI, with Mobileye acquiring Mentee Robotics for $900 million, Motional shifting to an “AI-first” AV development, and Chinese automakers gaining visibility while U.S. automakers were absent, as reported by TechCrunch. <a href="https://app.daily.dev/posts/MtQzzuuIN" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>New CLI tool integrates AI and developers with Figma</strong>: The <code>dannote/figma-use</code> command-line tool provides full read/write access to Figma, enabling AI agents and developers to programmatically create and modify designs using JSX syntax, optimized for token efficiency. <a href="https://app.daily.dev/posts/DfAoRLrdU" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Iran disrupts Starlink internet with military-grade GPS jamming</strong>: Iran successfully used military-grade GPS jammers to cut Starlink performance by up to 80% during a nationwide internet shutdown, raising concerns about Starlink’s reliability in conflict zones. <a href="https://app.daily.dev/posts/VVK9EFJ8w" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Open source models: cathedral, megachurch, and bazaar</strong>: Open source projects are categorized into three models: exclusive ‘cathedrals,’ large ‘megachurches’ like the Linux Foundation, and the majority ‘bazaar’ projects, which are often single-maintainer but critical to infrastructure and face significant funding challenges. <a href="https://app.daily.dev/posts/rr8Vo8FaG" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Claude pricing for individuals and organizations</strong>: Claude offers tiered pricing for individuals (Free, Pro, Max) and organizations (Team, Enterprise), with API costs based on tokens and additional charges for features like web search and code execution. <a href="https://app.daily.dev/posts/77ufsy3OX" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>FluentAI: low-code platform for LLM workflows</strong>: FluentAI is an open-source, low-code/no-code platform for visually building custom workflows with large language models in the browser, inspired by Node-RED and built with ReactFlow and Supabase. <a href="https://app.daily.dev/posts/T2Zh09ohU" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google’s Antigravity expands to compete with Anthropic’s Co-work</strong>: Google’s Antigravity is expanding beyond coding to include productivity features like creating presentations and organizing files directly from the IDE, leveraging Gemini 3 Flash to compete with Anthropic’s Co-work. <a href="https://app.daily.dev/posts/HVlI4ECqg" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI coding agents: the joy killers?</strong>: After 20 years of mastering Vim/Neovim, one developer is now questioning if learning fundamental tools like <code>awk</code> and <code>sed</code> is even worth it, thanks to AI coding agents. Apparently, these tools are “good enough” to automate the craft that once brought joy, leaving us to ponder what remains when the fun part of coding is outsourced to a chatbot. So much for the “art” of programming, I guess. <a href="https://app.daily.dev/posts/vFzyyNK3z" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Soft skills are the new hard skills, thanks to AI</strong>: Now that AI coding agents are apparently so good at writing code, software engineers are being told their “soft skills” like communication and specification are suddenly critical. The bottleneck has shifted from implementation to explaining what you actually want, because, let’s be honest, AI can’t read your mind (yet). So, if your AI-generated code is garbage, it’s probably your fault for not being articulate enough. <a href="https://app.daily.dev/posts/sifgyKGyd" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GitHub contribution graphs: still meaningless, apparently</strong>: Someone out there still thinks your GitHub contribution graph, LinkedIn profile, or LeetCode scores actually mean something. This article, however, reminds us that these metrics are still poor indicators of a developer’s true worth, especially when you consider things like private repos, different job markets, or, you know, actual human context. So, keep grinding those green squares if you want, but don’t expect them to tell your whole story. <a href="https://app.daily.dev/posts/WQGtqJsDu" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
