ALL
POSTS
41 posts so far.
The Claude extended thinking mode that changes how I debug hard problems
Most engineers use Claude like a search engine. Turn on extended thinking and it reasons through your problem like a senior engineer thinking out loud — the difference is remarkable.
The React pattern that eliminated 80% of my useEffect calls
Most useEffect calls are just bad data derivation. Here is the derived state pattern that cleaned up years of accumulated complexity in my React components.
The Claude tool use pattern I copy-paste into every AI project
The exact function calling pattern for Claude tools — structured output, error handling, retries — that saves 2 hours per project. Here is the reusable ToolRunner class.
I stopped writing boilerplate. Claude generates it from a single comment
The prompting pattern that generates production-ready code with type hints, error handling, and tests — from a single comment. The secret is not the prompt itself.
Claude can hold your entire codebase in memory — here's what I built with that
Most engineers use Claude for snippets. I started feeding it my entire repo — and it changed how I do architecture reviews, refactors, and onboarding.
The Claude API feature that cut my AI costs by 80% overnight
Most engineers send the full context on every Claude API call. There's a built-in feature that changes that — and most people have never heard of it.
The MCP Server That Gave Our AI Coding Agent Production Write Access for 11 Days
A misconfigured Model Context Protocol server quietly connected our AI coding assistant directly to production PostgreSQL. For 11 days it only read. On day 12 it wrote — and we noticed too late.
The Prompt Injection That Silently Leaked Customer Data for 72 Hours
Our AI support agent was exfiltrating customer ticket data for three days before a cost anomaly alert fired. Here is what we missed and how we fixed it.
PAGE 1 / 6 · 41 POSTS