Posts tagged "Engineering"

I pay for up to five $200/month AI subscriptions. The new M5 Mac Studio can hold frontier-class open weights in memory. Working out which one to buy corrected me twice.

I use AMICE to connect AI applications and models to the infrastructure, chips, and energy underneath them. Those dependencies help explain costs, delays, and limits.

An interactive field guide to model size, tokens, memory, speed, reasoning effort, open weights, deployment, routing, and the physical cost of an AI answer.

Matt Pocock's skills repo grew from eleven guardrail files into a full engineering process: a native Claude Code plugin, a planning layer that lives on your issue tracker, and a public folder of deprecated ideas.

Grow a silicon crystal, decode nanometers, compare CPUs and GPUs, open an HBM package, map the global supply chain, follow public money, and read chip export controls without calling every license a ban.

Follow an AI data center from power plant to utility, grid, fuel, cooling system, water source, contract, and unanswered reporting question.

My Cursor renewal forced an audit of every AI subscription I pay for. The math that settled it has nothing to do with price per token.

AI Coding Agents will sometimes make ridiculously large code changes. Over a million lines? Just let it be. It's probably fine.

Why understanding your product is the most critical skill for engineers using AI coding agents.

How the new /goal command changes Claude Code from giving up too easily to sticking with tough tasks.

An add-on that has made Claude Code more effective, more reliable, and lighter on my usage quota. Here is why I am keeping it.

AI coding agents accelerate codebase entropy. Matt Pocock's open-source Claude Code skills add the discipline your AI agent needs to write real software.

A quick look at recent AI releases including OpenAI's GPT-5.5, Anthropic's quality transparency, and xAI's massive Cursor acquisition rumors.

TrueUp is tracking over 67,000 open software engineering roles globally, the highest in more than three years. The headline is correct. What the number does not tell you is what those jobs actually require now.

The traditional lines between Product Owners and Product Managers are blurring as AI empowers engineers to build faster. Here is how responsibilities are shifting up the chain.

How the .claude folder turns your repo into a self-documenting workspace where your AI coding agent already knows the rules, the style, and the plan.

Cursor launched Composer-2 this week at a third the price of Composer 1.5. For UI and front-end work, it is hard to argue with the results.

Love Cursor's Ask mode and live in the terminal? Here's how to get the same codebase-aware answers from Claude Code CLI without the agent touching a single file.

Google quietly released a CLI that gives your terminal and your AI agent direct access to Gmail, Drive, Calendar, Docs, Sheets, and more. Agents can automate real work without a browser in the loop.

A quick look at a few things that happened in the AI space this week, including DoD negotiations, new free tiers, Cursor updates, and GPT 5.4.

A pragmatic 2026 stack for shipping real products with native iOS and Android code, Vercel backend, Clerk auth, Supabase data, and direct model APIs.

With a massive free tier upgrade to 50k MAU and new AI agent skills, Clerk has solved the biggest pain points of integrating application authentication.

Vercel plus AI coding agents and GitHub previews is the fastest way I know to ship apps, even if direct AWS could be cheaper at scale.

A recent study confirms what I've suspected: giving your AI coding agent too many instructions in repository-level context files actually reduces task success.

Claude Sonnet feels completely different in Cursor, Claude Code, and Antigravity. The toolchain is the product.

Anthropic and OpenAI released their flagship models within minutes of each other. For engineers who use these tools daily, the releases change what you can ship the same afternoon.

Anthropic's Super Bowl ad mocks ChatGPT for adding ads; the open question is whether ads can actually fix AI's economics. A look at the numbers.

Anthropic's recent study found AI users scored slightly lower on a coding quiz. The research has real signal, the methodology has limits, and you can still use AI without sacrificing learning.

OpenClaw is an open-source autonomous agent framework that crossed a psychological and technical line. It does not just recommend actions, it takes them.

For senior engineers, the most valuable thing an AI coding agent can do is read the system, not generate code. A guide to using Ask mode to understand complexity.

A practical breakdown of how Vercel, Supabase, and Clerk each own distinct responsibilities, integrate cleanly, and create better user-first product experiences.

Claude 4.5 is complete: how to split Opus, Sonnet, and Haiku across a Claude Code session so you get the most from the 5-hour quota.

The Model Context Protocol (MCP) is the latest trend in AI engineering. Overusing it is a fast track to context pollution and agent inefficiency.

A practical guide to using AI planning modes effectively. Stop treating your agent like a magic wand and start treating it like a junior engineer.

A deep dive into why passwords aren't enough and how the holy trinity of authentication protects your digital identity.

A quick reflection on how Anthropic's terminal agent has changed my day-to-day life as an engineer.

Cursor is ending its truly unlimited 'Auto' model usage on September 15, 2025. Here is how to lock in 365 days of infinite tokens before the window closes.

A new feature for technical posts giving you the 2 sentences you need to sound like you read the whole thing.

Anthropic's Claude Code CLI is shifting the development paradigm from manual editing to high-level orchestration. It is the most impressive tool in my stack because it runs where real engineering already happens.

Cursor represents a major shift in developer productivity. Notes on getting started and where it helps senior engineers most.

The search for the 'best' model is a distraction. Focus on the tools that actually solve your problems today.