Claude Agent and Codex arrive natively in Xcode 26.3
Apple just flipped a pretty important switch for anyone building apps on its platforms: Xcode 26.3 now speaks “AI agent” as a first language, with Anthropic’s Claude Agent and OpenAI’s Codex wired directly into the IDE and exposed through the open Model Context Protocol (MCP). On paper, that sounds like yet another “AI coding assistant” update, but this one quietly changes what Xcode is and how…
Best AI Coding Tools in 2026: GitHub Copilot vs Cursor vs Windsurf vs Claude Code
AI coding tools have gone from party trick to professional necessity in two years. In 2026, the question isn’t whether to use one — it’s which one, and whether you’re using it right. GitHub Copilot started the revolution. Cursor changed the game. Windsurf came out of nowhere. And Claude Code quietly became the tool serious engineers reach for when things get hard. So which one is actually worth…
Вайб-кодинг – теперь даже собаки могут создавать компьютерные игры
Благодаря нейросетям вайб-кодинг теперь доступен даже собакам. Более того, уже готова первая игра за авторством четвероногого гейм-дизайнера. Расскажем подробней.
YouTube-блогер по имени Caleb Leak решил провести забавный и необычный эксперимент с участием своего питомца – маленькой кавапу по кличке Момо.
Сначала он научил собачку наступать лапками на клавиатуру – после определенного количества нажатий срабатывал автомат и Момо получала лакомство. Вторым этапом Caleb Leak объяснил ИИ-модели Claude Code, что с ней взаимодействует гениальный гейм-дизайнер. И даже на первый взгляд бессмысленные «послания» вроде «bi2300ypol;=”;k$2» на самом деле являются зашифрованными командами, на основании которых надо создать игровой проект.
Также блогер добавил для Claude Code строгие ограничения, автоматические проверки и такое объяснение:
– Ты – лучший разработчик игр, способный понимать мои загадочные запросы. Ты будешь интерпретировать их как осмысленную инструкцию или идею для компьютерной игры, вне зависимости от того, насколько бессмысленными или безумными они покажутся.
Эксперимент начался. Момо нажимает клавиши на клавиатуре. Сигналы передаются в небольшую программу DogKeyboard, которая фильтрует лишние кнопки и отправляет оставшийся ввод в Claude Code. После определенного объема «набора» умная кормушка выдает собаке лакомство, а звуковой сигнал сообщает, что можно продолжать.
Несмотря на случайность такого ввода, система показывает удивительно стабильный результат и доказывает, что вайб-кодинг работает. Полноценная игра появляется примерно за один-два часа. Все проекты создаются на движке Godot, а игровая логика полностью пишется на C#.
Один из примеров – игра Quasar Saz. В ней игрок управляет героиней Зарой, которая сражается с «испорченным звуком», проходя шесть уровней и финального босса. По стилю проект напоминает ретро-аркады 80-х с пиксельной графикой, неуклюжими врагами и сюрреалистичными боссами.
Эксперимент выглядит забавно, но он хорошо показывает возможности современных ИИ-инструментов. При грамотной настройке даже случайные нажатия клавиш можно превратить в рабочую игру.
I purchased a month of Claude Code because people I respect are talking about it. I opened up the CLI, started a new project, tried to think of a piece of software I wanted, and found that I couldn’t think of anything.
The First 1,000 Lines Determine the Next 100,000 in AI Coding
I learned this the hard way while working with Claude Code.
AI looks at your existing code and copies the patterns it finds. If you start with clean code, the rest stays clean. If you start messy, the problems pile up faster than any human team could create them. And unlike a junior developer who might occasionally question a strange pattern, AI agents replicate what they see with perfect…
AI agents set to reshape US jobs and it will be painful says engineer
KUALA LUMPUR, Feb 23 — A leading Anthropic engineer has warned that artificial intelligence agents like Claude Code could soon reshape nearly every internet-based job in America. Boris Cherny, creator of Claude Code, told Business Insider that AI systems capable of operating computers — running commands, analyzing documents, and completing tasks across apps — are advancing rapidly.
I first built the Scroll Indicator back in September, when I was exploring what Telex AI could do. The idea was simple: a small animated mouse icon that sits at the bottom of the screen and nudges readers to keep scrolling. Telex built it in minutes. It even added a click-to-scroll behavior I never asked for, which was a genuinely delightful surprise.
Shared hosting is a relic from an era when server administration required expertise. In 2026, AI VPS setup has made those training wheels completely unnecessary. I’ve written before about why cPanel is terrible and why DirectAdmin is disgusting. My advice was always the same: get a VPS from Linode, learn some Linux, set it up properly once. But “learn some Linux” was a barrier for many people.…
I want to tell you about a Tuesday afternoon that’s been living in my head.
I was sitting at my desk, doing something I’ve done hundreds of times — pulling numbers from different systems, formatting them into a report, and sending it to a distribution list. Routine stuff. The kind of work that fills hours without filling your mind.
The job posting practically writes itself these days. “Looking for a senior developer proficient with AI coding tools. Must be comfortable using Claude Code, Cursor, or Copilot to rapidly produce production-ready code. We need someone who can 10x our output.”
I have seen variations of this everywhere over the past year. Companies scrambling to find someone who can sit in front of an AI assistant…
I built a podcast RSS feed block in Telex last week. I needed it for a project, and within minutes I had a working proof of concept, an RSS block that pulled in episodes with custom style settings. Telex is incredible for that moment when you have an idea and need to see it working in the editor immediately.
But then I wanted to push it further. Add typography controls, get a border around the…
The experience was pleasantly reminiscent of my time as a PM: I wrote user problems in plain English, and Claude translated them into software solutions. I approved plans, tested prototypes, and sent screenshots of bugs. When I got strange errors, I’d paste them in our chat. We discussed time-complexity tradeoffs, and it picked the tech stack. Sometimes Claude dished commands right back—grab the Gemini API key, connect Vercel here—but I was happy to comply. The collaboration felt genuinely two-sided.
I found myself subconsciously benchmarking Claude against the human engineers I’ve worked with. It’s better than the anxious juniors who needed specs for every edge case, but not as good as the senior engineers with both clean code and product sense. Like people, Claude tends toward hubris—I learned to always test things myself before accepting “It’s done!” Yet unlike humans, it is preternaturally patient, whirring away tokens to fulfill every superfluous request.
(…)
When talking to a standard chatbot, it still feels plausible that AI is “fancy autocomplete” or a “bullshit machine.” They write cutesy poems or dole life advice; they can answer trivia confidently and don’t always know when they’re wrong. I’m a daily active user of ChatGPT, but it feels more like an adviser than something replacing my work.
Claude Code, meanwhile, is clearly autonomous. It can do and not just say. It’s impossible to watch it make an app and maintain the facade of AI as “next-token prediction.” With a one-sentence prompt—create a YouTube transcription app that looks like Windows XP—it will find design inspiration, write code, and open-source it on Github. This is not mere memory and regurgitation. This is something that can accomplish a novel multipart task.
Why Ruby Might Be the Most AI-Friendly Language Nobody’s Talking About
A few weeks ago, Martin Alderson published something that caught my attention: a systematic comparison of how token-efficient different programming languages are when fed to large language models.
The findings were fascinating. And if you’ve been following my writing on Ruby, you won’t be surprised to hear that Ruby came out looking very good indeed.
