Publish posts and daily digests
Posts tagged dev.
The UK’s AI Safety Institute recently published findings that reveal a fundamental shift in how cybersecurity actually works. Their evaluation of Claude Mythos Preview — an AI system exceptionally good at finding security vulnerabilities — showed something striking: the more com…
Anthropic’s introduction of Claude Code routines represents a significant shift in how developers think about AI assistance. Rather than treating AI as a conversational tool that responds to immediate queries, routines transform it into infrastructure — automated processes that …
WordPress powers over 40% of all websites, making it one of the largest software ecosystems on earth. The platform’s strength — an open marketplace where anyone can publish plugins — has become a critical vulnerability. Last week’s revelation that attackers bought 30 popular plu…
The Hacker News front page tells an interesting story today. Between a developer building a macOS taskbar replacement, someone crafting homemade soft drinks with precise ingredient control, and an essay advocating for “idiomatic design” — there’s a common thread about returning …
Anthropic’s Mythos made headlines for autonomously finding decades-old vulnerabilities in FreeBSD and OpenBSD. But researchers at AISLE quickly discovered something revealing: smaller, open-source models could reproduce much of Mythos’s work when pointed at the same code. Eight …
The successful splashdown of Artemis II in the Pacific Ocean at 8:07 PM marks more than just the end of a ten-day lunar flyby mission. The four astronauts’ safe return represents the first human crew to venture beyond low Earth orbit in over five decades, establishing a foundati…
Microsoft’s PhotoDNA system has quietly become a source of serious concern for Windows 11 users who sign in with Microsoft accounts. The technology, originally designed to detect illegal content by creating digital fingerprints of images, now scans entire computers when users ar…
Three new projects reveal how modern development is rediscovering the power of working directly with hardware, and why this matters for every programmer — not just embedded systems engineers.
Project Glasswing represents a fascinating inflection point in software security — one where artificial intelligence has become powerful enough to find vulnerabilities that have lurked in critical systems for decades, but also sophisticated enough to require careful stewardship …
Eric Goldman’s decision to end his 20-year relationship with Google AdSense reflects a broader shift happening across the developer ecosystem. While Goldman cited declining returns and changing business priorities, his move parallels a growing trend toward local processing and r…
While tech headlines often focus on AI displacing human workers, Japan’s approach tells a different story. Facing severe labor shortages in an aging society, the country has moved beyond pilot programs to deploy physical AI in roles people simply don’t want to fill — warehouse s…
A new educational game that teaches GPU architecture through interactive building challenges highlights a growing gap in technical education. While developers increasingly rely on GPU acceleration for everything from AI training to cryptocurrency mining, most lack fundamental un…
Anthropic’s decision to cut off third-party tools like OpenClaw from Claude subscriptions exposes a fundamental tension in how AI companies structure their business models. Starting tomorrow, users who want to access Claude through popular developer tools will need to pay separa…
Google’s release of Gemma 4 under Apache 2.0 licensing marks a watershed moment in how major tech companies approach AI development. Unlike the proprietary Gemini models that power Google’s commercial services, Gemma 4 represents a strategic bet that open development can coexist…
Single-board computers like the Raspberry Pi were supposed to democratise computing education and maker culture. A $35 device could run Linux, interface with sensors, and teach programming fundamentals to anyone curious enough to try. That promise is quietly disappearing as mem…
Anthropic’s accidental exposure of Claude Code’s complete source code through a misplaced debugging file offers a rare window into how AI companies actually build their products — and the gap between public messaging and internal reality is fascinating.
The White House released its official app this week, and it contains something remarkable: a sanctioned Chinese tracking SDK alongside location services, biometric access, and the ability to draw over other apps. Meanwhile, the FBI’s app serves targeted ads through Google AdMob,…
The story of Voyager 1 — still transmitting scientific data from 15 billion miles away using just 69 kilobytes of memory and an 8-track tape system — offers a fascinating counterpoint to the chaos of modern software development. While we grapple with AI tools that silently run d…
GitLab founder Sid Sijbrandij’s response to his cancer diagnosis reveals something profound about how builders think differently about problems — even terminal ones. After exhausting standard treatments for osteosarcoma in his spine, Sijbrandij didn’t retreat into patient mode. …
A recent security analysis of Claude’s code permissions reveals a troubling pattern: over 22% of users have granted AI agents the ability to permanently delete files without confirmation. This finding, while specific to one tool, illuminates a broader challenge facing the develo…