ALL
POSTS
59 posts so far.
How I use Claude's 200K context window without burning my API budget
A 200K context window is powerful but expensive if you fill it carelessly. Here are the strategies I use to send the right context, not all the context.
My Claude code review workflow catches things I always miss manually
A structured prompt that uses Claude to review code for security issues, performance problems, and API design consistency — the things humans skip when tired.
I automated changelog generation with Claude and git log — never again manually
A script that turns raw git log into polished, user-facing changelogs grouped by type. Takes 30 seconds instead of 30 minutes before every release.
Claude's Batch API cuts my AI processing costs by 50% for overnight jobs
The Message Batches API processes up to 10,000 requests asynchronously at half the price. Perfect for document processing, classification, and content generation jobs.
Claude writes better API documentation than most humans do — here's the prompt
The prompt that turns a raw TypeScript function signature into complete API documentation with examples, error descriptions, and edge case notes.
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.
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.
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.
PAGE 2 / 3 · 18 POSTS